1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
13 input --> split ---------------------> overlay --> output
16 +-----> crop --> vflip -------+
19 This filtergraph splits the input stream in two streams, then sends one
20 stream through the crop filter and the vflip filter, before merging it
21 back with the other stream by overlaying it on top. You can use the
22 following command to achieve this:
25 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
28 The result will be that the top half of the video is mirrored
29 onto the bottom half of the output video.
31 Filters in the same linear chain are separated by commas, and distinct
32 linear chains of filters are separated by semicolons. In our example,
33 @var{crop,vflip} are in one linear chain, @var{split} and
34 @var{overlay} are separately in another. The points where the linear
35 chains join are labelled by names enclosed in square brackets. In the
36 example, the split filter generates two outputs that are associated to
37 the labels @var{[main]} and @var{[tmp]}.
39 The stream sent to the second output of @var{split}, labelled as
40 @var{[tmp]}, is processed through the @var{crop} filter, which crops
41 away the lower half part of the video, and then vertically flipped. The
42 @var{overlay} filter takes in input the first unchanged output of the
43 split filter (which was labelled as @var{[main]}), and overlay on its
44 lower half the output generated by the @var{crop,vflip} filterchain.
46 Some filters take in input a list of parameters: they are specified
47 after the filter name and an equal sign, and are separated from each other
50 There exist so-called @var{source filters} that do not have an
51 audio/video input, and @var{sink filters} that will not have audio/video
54 @c man end FILTERING INTRODUCTION
57 @c man begin GRAPH2DOT
59 The @file{graph2dot} program included in the FFmpeg @file{tools}
60 directory can be used to parse a filtergraph description and issue a
61 corresponding textual representation in the dot language.
68 to see how to use @file{graph2dot}.
70 You can then pass the dot description to the @file{dot} program (from
71 the graphviz suite of programs) and obtain a graphical representation
74 For example the sequence of commands:
76 echo @var{GRAPH_DESCRIPTION} | \
77 tools/graph2dot -o graph.tmp && \
78 dot -Tpng graph.tmp -o graph.png && \
82 can be used to create and display an image representing the graph
83 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
84 a complete self-contained graph, with its inputs and outputs explicitly defined.
85 For example if your command line is of the form:
87 ffmpeg -i infile -vf scale=640:360 outfile
89 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
91 nullsrc,scale=640:360,nullsink
93 you may also need to set the @var{nullsrc} parameters and add a @var{format}
94 filter in order to simulate a specific input file.
98 @chapter Filtergraph description
99 @c man begin FILTERGRAPH DESCRIPTION
101 A filtergraph is a directed graph of connected filters. It can contain
102 cycles, and there can be multiple links between a pair of
103 filters. Each link has one input pad on one side connecting it to one
104 filter from which it takes its input, and one output pad on the other
105 side connecting it to one filter accepting its output.
107 Each filter in a filtergraph is an instance of a filter class
108 registered in the application, which defines the features and the
109 number of input and output pads of the filter.
111 A filter with no input pads is called a "source", and a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph has a textual representation, which is recognized by the
118 @option{-filter}/@option{-vf}/@option{-af} and
119 @option{-filter_complex} options in @command{ffmpeg} and
120 @option{-vf}/@option{-af} in @command{ffplay}, and by the
121 @code{avfilter_graph_parse_ptr()} function defined in
122 @file{libavfilter/avfilter.h}.
124 A filterchain consists of a sequence of connected filters, each one
125 connected to the previous one in the sequence. A filterchain is
126 represented by a list of ","-separated filter descriptions.
128 A filtergraph consists of a sequence of filterchains. A sequence of
129 filterchains is represented by a list of ";"-separated filterchain
132 A filter is represented by a string of the form:
133 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
135 @var{filter_name} is the name of the filter class of which the
136 described filter is an instance of, and has to be the name of one of
137 the filter classes registered in the program.
138 The name of the filter class is optionally followed by a string
141 @var{arguments} is a string which contains the parameters used to
142 initialize the filter instance. It may have one of two forms:
146 A ':'-separated list of @var{key=value} pairs.
149 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
150 the option names in the order they are declared. E.g. the @code{fade} filter
151 declares three options in this order -- @option{type}, @option{start_frame} and
152 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
153 @var{in} is assigned to the option @option{type}, @var{0} to
154 @option{start_frame} and @var{30} to @option{nb_frames}.
157 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
158 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
159 follow the same constraints order of the previous point. The following
160 @var{key=value} pairs can be set in any preferred order.
164 If the option value itself is a list of items (e.g. the @code{format} filter
165 takes a list of pixel formats), the items in the list are usually separated by
168 The list of arguments can be quoted using the character @samp{'} as initial
169 and ending mark, and the character @samp{\} for escaping the characters
170 within the quoted text; otherwise the argument string is considered
171 terminated when the next special character (belonging to the set
172 @samp{[]=;,}) is encountered.
174 The name and arguments of the filter are optionally preceded and
175 followed by a list of link labels.
176 A link label allows one to name a link and associate it to a filter output
177 or input pad. The preceding labels @var{in_link_1}
178 ... @var{in_link_N}, are associated to the filter input pads,
179 the following labels @var{out_link_1} ... @var{out_link_M}, are
180 associated to the output pads.
182 When two link labels with the same name are found in the
183 filtergraph, a link between the corresponding input and output pad is
186 If an output pad is not labelled, it is linked by default to the first
187 unlabelled input pad of the next filter in the filterchain.
188 For example in the filterchain
190 nullsrc, split[L1], [L2]overlay, nullsink
192 the split filter instance has two output pads, and the overlay filter
193 instance two input pads. The first output pad of split is labelled
194 "L1", the first input pad of overlay is labelled "L2", and the second
195 output pad of split is linked to the second input pad of overlay,
196 which are both unlabelled.
198 In a filter description, if the input label of the first filter is not
199 specified, "in" is assumed; if the output label of the last filter is not
200 specified, "out" is assumed.
202 In a complete filterchain all the unlabelled filter input and output
203 pads must be connected. A filtergraph is considered valid if all the
204 filter input and output pads of all the filterchains are connected.
206 Libavfilter will automatically insert @ref{scale} filters where format
207 conversion is required. It is possible to specify swscale flags
208 for those automatically inserted scalers by prepending
209 @code{sws_flags=@var{flags};}
210 to the filtergraph description.
212 Here is a BNF description of the filtergraph syntax:
214 @var{NAME} ::= sequence of alphanumeric characters and '_'
215 @var{LINKLABEL} ::= "[" @var{NAME} "]"
216 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
217 @var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
218 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
219 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
220 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
223 @section Notes on filtergraph escaping
225 Filtergraph description composition entails several levels of
226 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
227 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
228 information about the employed escaping procedure.
230 A first level escaping affects the content of each filter option
231 value, which may contain the special character @code{:} used to
232 separate values, or one of the escaping characters @code{\'}.
234 A second level escaping affects the whole filter description, which
235 may contain the escaping characters @code{\'} or the special
236 characters @code{[],;} used by the filtergraph description.
238 Finally, when you specify a filtergraph on a shell commandline, you
239 need to perform a third level escaping for the shell special
240 characters contained within it.
242 For example, consider the following string to be embedded in
243 the @ref{drawtext} filter description @option{text} value:
245 this is a 'string': may contain one, or more, special characters
248 This string contains the @code{'} special escaping character, and the
249 @code{:} special character, so it needs to be escaped in this way:
251 text=this is a \'string\'\: may contain one, or more, special characters
254 A second level of escaping is required when embedding the filter
255 description in a filtergraph description, in order to escape all the
256 filtergraph special characters. Thus the example above becomes:
258 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
260 (note that in addition to the @code{\'} escaping special characters,
261 also @code{,} needs to be escaped).
263 Finally an additional level of escaping is needed when writing the
264 filtergraph description in a shell command, which depends on the
265 escaping rules of the adopted shell. For example, assuming that
266 @code{\} is special and needs to be escaped with another @code{\}, the
267 previous string will finally result in:
269 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
272 @chapter Timeline editing
274 Some filters support a generic @option{enable} option. For the filters
275 supporting timeline editing, this option can be set to an expression which is
276 evaluated before sending a frame to the filter. If the evaluation is non-zero,
277 the filter will be enabled, otherwise the frame will be sent unchanged to the
278 next filter in the filtergraph.
280 The expression accepts the following values:
283 timestamp expressed in seconds, NAN if the input timestamp is unknown
286 sequential number of the input frame, starting from 0
289 the position in the file of the input frame, NAN if unknown
293 width and height of the input frame if video
296 Additionally, these filters support an @option{enable} command that can be used
297 to re-define the expression.
299 Like any other filtering option, the @option{enable} option follows the same
302 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
303 minutes, and a @ref{curves} filter starting at 3 seconds:
305 smartblur = enable='between(t,10,3*60)',
306 curves = enable='gte(t,3)' : preset=cross_process
309 See @code{ffmpeg -filters} to view which filters have timeline support.
311 @c man end FILTERGRAPH DESCRIPTION
313 @chapter Audio Filters
314 @c man begin AUDIO FILTERS
316 When you configure your FFmpeg build, you can disable any of the
317 existing filters using @code{--disable-filters}.
318 The configure output will show the audio filters included in your
321 Below is a description of the currently available audio filters.
325 A compressor is mainly used to reduce the dynamic range of a signal.
326 Especially modern music is mostly compressed at a high ratio to
327 improve the overall loudness. It's done to get the highest attention
328 of a listener, "fatten" the sound and bring more "power" to the track.
329 If a signal is compressed too much it may sound dull or "dead"
330 afterwards or it may start to "pump" (which could be a powerful effect
331 but can also destroy a track completely).
332 The right compression is the key to reach a professional sound and is
333 the high art of mixing and mastering. Because of its complex settings
334 it may take a long time to get the right feeling for this kind of effect.
336 Compression is done by detecting the volume above a chosen level
337 @code{threshold} and dividing it by the factor set with @code{ratio}.
338 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
339 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
340 the signal would cause distortion of the waveform the reduction can be
341 levelled over the time. This is done by setting "Attack" and "Release".
342 @code{attack} determines how long the signal has to rise above the threshold
343 before any reduction will occur and @code{release} sets the time the signal
344 has to fall below the threshold to reduce the reduction again. Shorter signals
345 than the chosen attack time will be left untouched.
346 The overall reduction of the signal can be made up afterwards with the
347 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
348 raising the makeup to this level results in a signal twice as loud than the
349 source. To gain a softer entry in the compression the @code{knee} flattens the
350 hard edge at the threshold in the range of the chosen decibels.
352 The filter accepts the following options:
356 Set input gain. Default is 1. Range is between 0.015625 and 64.
359 If a signal of second stream rises above this level it will affect the gain
360 reduction of the first stream.
361 By default it is 0.125. Range is between 0.00097563 and 1.
364 Set a ratio by which the signal is reduced. 1:2 means that if the level
365 rose 4dB above the threshold, it will be only 2dB above after the reduction.
366 Default is 2. Range is between 1 and 20.
369 Amount of milliseconds the signal has to rise above the threshold before gain
370 reduction starts. Default is 20. Range is between 0.01 and 2000.
373 Amount of milliseconds the signal has to fall below the threshold before
374 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
377 Set the amount by how much signal will be amplified after processing.
378 Default is 2. Range is from 1 and 64.
381 Curve the sharp knee around the threshold to enter gain reduction more softly.
382 Default is 2.82843. Range is between 1 and 8.
385 Choose if the @code{average} level between all channels of input stream
386 or the louder(@code{maximum}) channel of input stream affects the
387 reduction. Default is @code{average}.
390 Should the exact signal be taken in case of @code{peak} or an RMS one in case
391 of @code{rms}. Default is @code{rms} which is mostly smoother.
394 How much to use compressed signal in output. Default is 1.
395 Range is between 0 and 1.
400 Apply cross fade from one input audio stream to another input audio stream.
401 The cross fade is applied for specified duration near the end of first stream.
403 The filter accepts the following options:
407 Specify the number of samples for which the cross fade effect has to last.
408 At the end of the cross fade effect the first input audio will be completely
409 silent. Default is 44100.
412 Specify the duration of the cross fade effect. See
413 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
414 for the accepted syntax.
415 By default the duration is determined by @var{nb_samples}.
416 If set this option is used instead of @var{nb_samples}.
419 Should first stream end overlap with second stream start. Default is enabled.
422 Set curve for cross fade transition for first stream.
425 Set curve for cross fade transition for second stream.
427 For description of available curve types see @ref{afade} filter description.
434 Cross fade from one input to another:
436 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
440 Cross fade from one input to another but without overlapping:
442 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
448 Reduce audio bit resolution.
450 This filter is bit crusher with enhanced functionality. A bit crusher
451 is used to audibly reduce number of bits an audio signal is sampled
452 with. This doesn't change the bit depth at all, it just produces the
453 effect. Material reduced in bit depth sounds more harsh and "digital".
454 This filter is able to even round to continuous values instead of discrete
456 Additionally it has a D/C offset which results in different crushing of
457 the lower and the upper half of the signal.
458 An Anti-Aliasing setting is able to produce "softer" crushing sounds.
460 Another feature of this filter is the logarithmic mode.
461 This setting switches from linear distances between bits to logarithmic ones.
462 The result is a much more "natural" sounding crusher which doesn't gate low
463 signals for example. The human ear has a logarithmic perception, too
464 so this kind of crushing is much more pleasant.
465 Logarithmic crushing is also able to get anti-aliased.
467 The filter accepts the following options:
483 Can be linear: @code{lin} or logarithmic: @code{log}.
492 Set sample reduction.
495 Enable LFO. By default disabled.
506 Delay one or more audio channels.
508 Samples in delayed channel are filled with silence.
510 The filter accepts the following option:
514 Set list of delays in milliseconds for each channel separated by '|'.
515 At least one delay greater than 0 should be provided.
516 Unused delays will be silently ignored. If number of given delays is
517 smaller than number of channels all remaining channels will not be delayed.
518 If you want to delay exact number of samples, append 'S' to number.
525 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
526 the second channel (and any other channels that may be present) unchanged.
532 Delay second channel by 500 samples, the third channel by 700 samples and leave
533 the first channel (and any other channels that may be present) unchanged.
541 Apply echoing to the input audio.
543 Echoes are reflected sound and can occur naturally amongst mountains
544 (and sometimes large buildings) when talking or shouting; digital echo
545 effects emulate this behaviour and are often used to help fill out the
546 sound of a single instrument or vocal. The time difference between the
547 original signal and the reflection is the @code{delay}, and the
548 loudness of the reflected signal is the @code{decay}.
549 Multiple echoes can have different delays and decays.
551 A description of the accepted parameters follows.
555 Set input gain of reflected signal. Default is @code{0.6}.
558 Set output gain of reflected signal. Default is @code{0.3}.
561 Set list of time intervals in milliseconds between original signal and reflections
562 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
563 Default is @code{1000}.
566 Set list of loudnesses of reflected signals separated by '|'.
567 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
568 Default is @code{0.5}.
575 Make it sound as if there are twice as many instruments as are actually playing:
577 aecho=0.8:0.88:60:0.4
581 If delay is very short, then it sound like a (metallic) robot playing music:
587 A longer delay will sound like an open air concert in the mountains:
589 aecho=0.8:0.9:1000:0.3
593 Same as above but with one more mountain:
595 aecho=0.8:0.9:1000|1800:0.3|0.25
600 Audio emphasis filter creates or restores material directly taken from LPs or
601 emphased CDs with different filter curves. E.g. to store music on vinyl the
602 signal has to be altered by a filter first to even out the disadvantages of
603 this recording medium.
604 Once the material is played back the inverse filter has to be applied to
605 restore the distortion of the frequency response.
607 The filter accepts the following options:
617 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
618 use @code{production} mode. Default is @code{reproduction} mode.
621 Set filter type. Selects medium. Can be one of the following:
633 select Compact Disc (CD).
639 select 50µs (FM-KF).
641 select 75µs (FM-KF).
647 Modify an audio signal according to the specified expressions.
649 This filter accepts one or more expressions (one for each channel),
650 which are evaluated and used to modify a corresponding audio signal.
652 It accepts the following parameters:
656 Set the '|'-separated expressions list for each separate channel. If
657 the number of input channels is greater than the number of
658 expressions, the last specified expression is used for the remaining
661 @item channel_layout, c
662 Set output channel layout. If not specified, the channel layout is
663 specified by the number of expressions. If set to @samp{same}, it will
664 use by default the same input channel layout.
667 Each expression in @var{exprs} can contain the following constants and functions:
671 channel number of the current expression
674 number of the evaluated sample, starting from 0
680 time of the evaluated sample expressed in seconds
683 @item nb_out_channels
684 input and output number of channels
687 the value of input channel with number @var{CH}
690 Note: this filter is slow. For faster processing you should use a
699 aeval=val(ch)/2:c=same
703 Invert phase of the second channel:
712 Apply fade-in/out effect to input audio.
714 A description of the accepted parameters follows.
718 Specify the effect type, can be either @code{in} for fade-in, or
719 @code{out} for a fade-out effect. Default is @code{in}.
721 @item start_sample, ss
722 Specify the number of the start sample for starting to apply the fade
723 effect. Default is 0.
726 Specify the number of samples for which the fade effect has to last. At
727 the end of the fade-in effect the output audio will have the same
728 volume as the input audio, at the end of the fade-out transition
729 the output audio will be silence. Default is 44100.
732 Specify the start time of the fade effect. Default is 0.
733 The value must be specified as a time duration; see
734 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
735 for the accepted syntax.
736 If set this option is used instead of @var{start_sample}.
739 Specify the duration of the fade effect. See
740 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
741 for the accepted syntax.
742 At the end of the fade-in effect the output audio will have the same
743 volume as the input audio, at the end of the fade-out transition
744 the output audio will be silence.
745 By default the duration is determined by @var{nb_samples}.
746 If set this option is used instead of @var{nb_samples}.
749 Set curve for fade transition.
751 It accepts the following values:
754 select triangular, linear slope (default)
756 select quarter of sine wave
758 select half of sine wave
760 select exponential sine wave
764 select inverted parabola
778 select inverted quarter of sine wave
780 select inverted half of sine wave
782 select double-exponential seat
784 select double-exponential sigmoid
792 Fade in first 15 seconds of audio:
798 Fade out last 25 seconds of a 900 seconds audio:
800 afade=t=out:st=875:d=25
805 Apply arbitrary expressions to samples in frequency domain.
809 Set frequency domain real expression for each separate channel separated
810 by '|'. Default is "1".
811 If the number of input channels is greater than the number of
812 expressions, the last specified expression is used for the remaining
816 Set frequency domain imaginary expression for each separate channel
817 separated by '|'. If not set, @var{real} option is used.
819 Each expression in @var{real} and @var{imag} can contain the following
827 current frequency bin number
830 number of available bins
833 channel number of the current expression
845 It accepts the following values:
861 Default is @code{w4096}
864 Set window function. Default is @code{hann}.
867 Set window overlap. If set to 1, the recommended overlap for selected
868 window function will be picked. Default is @code{0.75}.
875 Leave almost only low frequencies in audio:
877 afftfilt="1-clip((b/nb)*b,0,1)"
884 Set output format constraints for the input audio. The framework will
885 negotiate the most appropriate format to minimize conversions.
887 It accepts the following parameters:
891 A '|'-separated list of requested sample formats.
894 A '|'-separated list of requested sample rates.
896 @item channel_layouts
897 A '|'-separated list of requested channel layouts.
899 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
900 for the required syntax.
903 If a parameter is omitted, all values are allowed.
905 Force the output to either unsigned 8-bit or signed 16-bit stereo
907 aformat=sample_fmts=u8|s16:channel_layouts=stereo
912 A gate is mainly used to reduce lower parts of a signal. This kind of signal
913 processing reduces disturbing noise between useful signals.
915 Gating is done by detecting the volume below a chosen level @var{threshold}
916 and dividing it by the factor set with @var{ratio}. The bottom of the noise
917 floor is set via @var{range}. Because an exact manipulation of the signal
918 would cause distortion of the waveform the reduction can be levelled over
919 time. This is done by setting @var{attack} and @var{release}.
921 @var{attack} determines how long the signal has to fall below the threshold
922 before any reduction will occur and @var{release} sets the time the signal
923 has to rise above the threshold to reduce the reduction again.
924 Shorter signals than the chosen attack time will be left untouched.
928 Set input level before filtering.
929 Default is 1. Allowed range is from 0.015625 to 64.
932 Set the level of gain reduction when the signal is below the threshold.
933 Default is 0.06125. Allowed range is from 0 to 1.
936 If a signal rises above this level the gain reduction is released.
937 Default is 0.125. Allowed range is from 0 to 1.
940 Set a ratio by which the signal is reduced.
941 Default is 2. Allowed range is from 1 to 9000.
944 Amount of milliseconds the signal has to rise above the threshold before gain
946 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
949 Amount of milliseconds the signal has to fall below the threshold before the
950 reduction is increased again. Default is 250 milliseconds.
951 Allowed range is from 0.01 to 9000.
954 Set amount of amplification of signal after processing.
955 Default is 1. Allowed range is from 1 to 64.
958 Curve the sharp knee around the threshold to enter gain reduction more softly.
959 Default is 2.828427125. Allowed range is from 1 to 8.
962 Choose if exact signal should be taken for detection or an RMS like one.
963 Default is @code{rms}. Can be @code{peak} or @code{rms}.
966 Choose if the average level between all channels or the louder channel affects
968 Default is @code{average}. Can be @code{average} or @code{maximum}.
973 The limiter prevents an input signal from rising over a desired threshold.
974 This limiter uses lookahead technology to prevent your signal from distorting.
975 It means that there is a small delay after the signal is processed. Keep in mind
976 that the delay it produces is the attack time you set.
978 The filter accepts the following options:
982 Set input gain. Default is 1.
985 Set output gain. Default is 1.
988 Don't let signals above this level pass the limiter. Default is 1.
991 The limiter will reach its attenuation level in this amount of time in
992 milliseconds. Default is 5 milliseconds.
995 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
996 Default is 50 milliseconds.
999 When gain reduction is always needed ASC takes care of releasing to an
1000 average reduction level rather than reaching a reduction of 0 in the release
1004 Select how much the release time is affected by ASC, 0 means nearly no changes
1005 in release time while 1 produces higher release times.
1008 Auto level output signal. Default is enabled.
1009 This normalizes audio back to 0dB if enabled.
1012 Depending on picked setting it is recommended to upsample input 2x or 4x times
1013 with @ref{aresample} before applying this filter.
1017 Apply a two-pole all-pass filter with central frequency (in Hz)
1018 @var{frequency}, and filter-width @var{width}.
1019 An all-pass filter changes the audio's frequency to phase relationship
1020 without changing its frequency to amplitude relationship.
1022 The filter accepts the following options:
1026 Set frequency in Hz.
1029 Set method to specify band-width of filter.
1042 Specify the band-width of a filter in width_type units.
1045 Specify which channels to filter, by default all available are filtered.
1052 The filter accepts the following options:
1056 Set the number of loops.
1059 Set maximal number of samples.
1062 Set first sample of loop.
1068 Merge two or more audio streams into a single multi-channel stream.
1070 The filter accepts the following options:
1075 Set the number of inputs. Default is 2.
1079 If the channel layouts of the inputs are disjoint, and therefore compatible,
1080 the channel layout of the output will be set accordingly and the channels
1081 will be reordered as necessary. If the channel layouts of the inputs are not
1082 disjoint, the output will have all the channels of the first input then all
1083 the channels of the second input, in that order, and the channel layout of
1084 the output will be the default value corresponding to the total number of
1087 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1088 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1089 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1090 first input, b1 is the first channel of the second input).
1092 On the other hand, if both input are in stereo, the output channels will be
1093 in the default order: a1, a2, b1, b2, and the channel layout will be
1094 arbitrarily set to 4.0, which may or may not be the expected value.
1096 All inputs must have the same sample rate, and format.
1098 If inputs do not have the same duration, the output will stop with the
1101 @subsection Examples
1105 Merge two mono files into a stereo stream:
1107 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1111 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1113 ffmpeg -i input.mkv -filter_complex "[0:1][0:2][0:3][0:4][0:5][0:6] amerge=inputs=6" -c:a pcm_s16le output.mkv
1119 Mixes multiple audio inputs into a single output.
1121 Note that this filter only supports float samples (the @var{amerge}
1122 and @var{pan} audio filters support many formats). If the @var{amix}
1123 input has integer samples then @ref{aresample} will be automatically
1124 inserted to perform the conversion to float samples.
1128 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1130 will mix 3 input audio streams to a single output with the same duration as the
1131 first input and a dropout transition time of 3 seconds.
1133 It accepts the following parameters:
1137 The number of inputs. If unspecified, it defaults to 2.
1140 How to determine the end-of-stream.
1144 The duration of the longest input. (default)
1147 The duration of the shortest input.
1150 The duration of the first input.
1154 @item dropout_transition
1155 The transition time, in seconds, for volume renormalization when an input
1156 stream ends. The default value is 2 seconds.
1160 @section anequalizer
1162 High-order parametric multiband equalizer for each channel.
1164 It accepts the following parameters:
1168 This option string is in format:
1169 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1170 Each equalizer band is separated by '|'.
1174 Set channel number to which equalization will be applied.
1175 If input doesn't have that channel the entry is ignored.
1178 Set central frequency for band.
1179 If input doesn't have that frequency the entry is ignored.
1182 Set band width in hertz.
1185 Set band gain in dB.
1188 Set filter type for band, optional, can be:
1192 Butterworth, this is default.
1203 With this option activated frequency response of anequalizer is displayed
1207 Set video stream size. Only useful if curves option is activated.
1210 Set max gain that will be displayed. Only useful if curves option is activated.
1211 Setting this to a reasonable value makes it possible to display gain which is derived from
1212 neighbour bands which are too close to each other and thus produce higher gain
1213 when both are activated.
1216 Set frequency scale used to draw frequency response in video output.
1217 Can be linear or logarithmic. Default is logarithmic.
1220 Set color for each channel curve which is going to be displayed in video stream.
1221 This is list of color names separated by space or by '|'.
1222 Unrecognised or missing colors will be replaced by white color.
1225 @subsection Examples
1229 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1230 for first 2 channels using Chebyshev type 1 filter:
1232 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1236 @subsection Commands
1238 This filter supports the following commands:
1241 Alter existing filter parameters.
1242 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1244 @var{fN} is existing filter number, starting from 0, if no such filter is available
1246 @var{freq} set new frequency parameter.
1247 @var{width} set new width parameter in herz.
1248 @var{gain} set new gain parameter in dB.
1250 Full filter invocation with asendcmd may look like this:
1251 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1256 Pass the audio source unchanged to the output.
1260 Pad the end of an audio stream with silence.
1262 This can be used together with @command{ffmpeg} @option{-shortest} to
1263 extend audio streams to the same length as the video stream.
1265 A description of the accepted options follows.
1269 Set silence packet size. Default value is 4096.
1272 Set the number of samples of silence to add to the end. After the
1273 value is reached, the stream is terminated. This option is mutually
1274 exclusive with @option{whole_len}.
1277 Set the minimum total number of samples in the output audio stream. If
1278 the value is longer than the input audio length, silence is added to
1279 the end, until the value is reached. This option is mutually exclusive
1280 with @option{pad_len}.
1283 If neither the @option{pad_len} nor the @option{whole_len} option is
1284 set, the filter will add silence to the end of the input stream
1287 @subsection Examples
1291 Add 1024 samples of silence to the end of the input:
1297 Make sure the audio output will contain at least 10000 samples, pad
1298 the input with silence if required:
1300 apad=whole_len=10000
1304 Use @command{ffmpeg} to pad the audio input with silence, so that the
1305 video stream will always result the shortest and will be converted
1306 until the end in the output file when using the @option{shortest}
1309 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1314 Add a phasing effect to the input audio.
1316 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1317 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1319 A description of the accepted parameters follows.
1323 Set input gain. Default is 0.4.
1326 Set output gain. Default is 0.74
1329 Set delay in milliseconds. Default is 3.0.
1332 Set decay. Default is 0.4.
1335 Set modulation speed in Hz. Default is 0.5.
1338 Set modulation type. Default is triangular.
1340 It accepts the following values:
1349 Audio pulsator is something between an autopanner and a tremolo.
1350 But it can produce funny stereo effects as well. Pulsator changes the volume
1351 of the left and right channel based on a LFO (low frequency oscillator) with
1352 different waveforms and shifted phases.
1353 This filter have the ability to define an offset between left and right
1354 channel. An offset of 0 means that both LFO shapes match each other.
1355 The left and right channel are altered equally - a conventional tremolo.
1356 An offset of 50% means that the shape of the right channel is exactly shifted
1357 in phase (or moved backwards about half of the frequency) - pulsator acts as
1358 an autopanner. At 1 both curves match again. Every setting in between moves the
1359 phase shift gapless between all stages and produces some "bypassing" sounds with
1360 sine and triangle waveforms. The more you set the offset near 1 (starting from
1361 the 0.5) the faster the signal passes from the left to the right speaker.
1363 The filter accepts the following options:
1367 Set input gain. By default it is 1. Range is [0.015625 - 64].
1370 Set output gain. By default it is 1. Range is [0.015625 - 64].
1373 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1374 sawup or sawdown. Default is sine.
1377 Set modulation. Define how much of original signal is affected by the LFO.
1380 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1383 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1386 Set pulse width. Default is 1. Allowed range is [0 - 2].
1389 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1392 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1396 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1400 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1401 if timing is set to hz.
1407 Resample the input audio to the specified parameters, using the
1408 libswresample library. If none are specified then the filter will
1409 automatically convert between its input and output.
1411 This filter is also able to stretch/squeeze the audio data to make it match
1412 the timestamps or to inject silence / cut out audio to make it match the
1413 timestamps, do a combination of both or do neither.
1415 The filter accepts the syntax
1416 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1417 expresses a sample rate and @var{resampler_options} is a list of
1418 @var{key}=@var{value} pairs, separated by ":". See the
1419 @ref{Resampler Options,,the "Resampler Options" section in the
1420 ffmpeg-resampler(1) manual,ffmpeg-resampler}
1421 for the complete list of supported options.
1423 @subsection Examples
1427 Resample the input audio to 44100Hz:
1433 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1434 samples per second compensation:
1436 aresample=async=1000
1442 Reverse an audio clip.
1444 Warning: This filter requires memory to buffer the entire clip, so trimming
1447 @subsection Examples
1451 Take the first 5 seconds of a clip, and reverse it.
1453 atrim=end=5,areverse
1457 @section asetnsamples
1459 Set the number of samples per each output audio frame.
1461 The last output packet may contain a different number of samples, as
1462 the filter will flush all the remaining samples when the input audio
1465 The filter accepts the following options:
1469 @item nb_out_samples, n
1470 Set the number of frames per each output audio frame. The number is
1471 intended as the number of samples @emph{per each channel}.
1472 Default value is 1024.
1475 If set to 1, the filter will pad the last audio frame with zeroes, so
1476 that the last frame will contain the same number of samples as the
1477 previous ones. Default value is 1.
1480 For example, to set the number of per-frame samples to 1234 and
1481 disable padding for the last frame, use:
1483 asetnsamples=n=1234:p=0
1488 Set the sample rate without altering the PCM data.
1489 This will result in a change of speed and pitch.
1491 The filter accepts the following options:
1494 @item sample_rate, r
1495 Set the output sample rate. Default is 44100 Hz.
1500 Show a line containing various information for each input audio frame.
1501 The input audio is not modified.
1503 The shown line contains a sequence of key/value pairs of the form
1504 @var{key}:@var{value}.
1506 The following values are shown in the output:
1510 The (sequential) number of the input frame, starting from 0.
1513 The presentation timestamp of the input frame, in time base units; the time base
1514 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1517 The presentation timestamp of the input frame in seconds.
1520 position of the frame in the input stream, -1 if this information in
1521 unavailable and/or meaningless (for example in case of synthetic audio)
1530 The sample rate for the audio frame.
1533 The number of samples (per channel) in the frame.
1536 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1537 audio, the data is treated as if all the planes were concatenated.
1539 @item plane_checksums
1540 A list of Adler-32 checksums for each data plane.
1546 Display time domain statistical information about the audio channels.
1547 Statistics are calculated and displayed for each audio channel and,
1548 where applicable, an overall figure is also given.
1550 It accepts the following option:
1553 Short window length in seconds, used for peak and trough RMS measurement.
1554 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1558 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1559 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1562 Available keys for each channel are:
1593 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1594 this @code{lavfi.astats.Overall.Peak_count}.
1596 For description what each key means read below.
1599 Set number of frame after which stats are going to be recalculated.
1600 Default is disabled.
1603 A description of each shown parameter follows:
1607 Mean amplitude displacement from zero.
1610 Minimal sample level.
1613 Maximal sample level.
1615 @item Min difference
1616 Minimal difference between two consecutive samples.
1618 @item Max difference
1619 Maximal difference between two consecutive samples.
1621 @item Mean difference
1622 Mean difference between two consecutive samples.
1623 The average of each difference between two consecutive samples.
1627 Standard peak and RMS level measured in dBFS.
1631 Peak and trough values for RMS level measured over a short window.
1634 Standard ratio of peak to RMS level (note: not in dB).
1637 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1638 (i.e. either @var{Min level} or @var{Max level}).
1641 Number of occasions (not the number of samples) that the signal attained either
1642 @var{Min level} or @var{Max level}.
1645 Overall bit depth of audio. Number of bits used for each sample.
1652 The filter accepts exactly one parameter, the audio tempo. If not
1653 specified then the filter will assume nominal 1.0 tempo. Tempo must
1654 be in the [0.5, 2.0] range.
1656 @subsection Examples
1660 Slow down audio to 80% tempo:
1666 To speed up audio to 125% tempo:
1674 Trim the input so that the output contains one continuous subpart of the input.
1676 It accepts the following parameters:
1679 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1680 sample with the timestamp @var{start} will be the first sample in the output.
1683 Specify time of the first audio sample that will be dropped, i.e. the
1684 audio sample immediately preceding the one with the timestamp @var{end} will be
1685 the last sample in the output.
1688 Same as @var{start}, except this option sets the start timestamp in samples
1692 Same as @var{end}, except this option sets the end timestamp in samples instead
1696 The maximum duration of the output in seconds.
1699 The number of the first sample that should be output.
1702 The number of the first sample that should be dropped.
1705 @option{start}, @option{end}, and @option{duration} are expressed as time
1706 duration specifications; see
1707 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1709 Note that the first two sets of the start/end options and the @option{duration}
1710 option look at the frame timestamp, while the _sample options simply count the
1711 samples that pass through the filter. So start/end_pts and start/end_sample will
1712 give different results when the timestamps are wrong, inexact or do not start at
1713 zero. Also note that this filter does not modify the timestamps. If you wish
1714 to have the output timestamps start at zero, insert the asetpts filter after the
1717 If multiple start or end options are set, this filter tries to be greedy and
1718 keep all samples that match at least one of the specified constraints. To keep
1719 only the part that matches all the constraints at once, chain multiple atrim
1722 The defaults are such that all the input is kept. So it is possible to set e.g.
1723 just the end values to keep everything before the specified time.
1728 Drop everything except the second minute of input:
1730 ffmpeg -i INPUT -af atrim=60:120
1734 Keep only the first 1000 samples:
1736 ffmpeg -i INPUT -af atrim=end_sample=1000
1743 Apply a two-pole Butterworth band-pass filter with central
1744 frequency @var{frequency}, and (3dB-point) band-width width.
1745 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1746 instead of the default: constant 0dB peak gain.
1747 The filter roll off at 6dB per octave (20dB per decade).
1749 The filter accepts the following options:
1753 Set the filter's central frequency. Default is @code{3000}.
1756 Constant skirt gain if set to 1. Defaults to 0.
1759 Set method to specify band-width of filter.
1772 Specify the band-width of a filter in width_type units.
1775 Specify which channels to filter, by default all available are filtered.
1780 Apply a two-pole Butterworth band-reject filter with central
1781 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1782 The filter roll off at 6dB per octave (20dB per decade).
1784 The filter accepts the following options:
1788 Set the filter's central frequency. Default is @code{3000}.
1791 Set method to specify band-width of filter.
1804 Specify the band-width of a filter in width_type units.
1807 Specify which channels to filter, by default all available are filtered.
1812 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1813 shelving filter with a response similar to that of a standard
1814 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1816 The filter accepts the following options:
1820 Give the gain at 0 Hz. Its useful range is about -20
1821 (for a large cut) to +20 (for a large boost).
1822 Beware of clipping when using a positive gain.
1825 Set the filter's central frequency and so can be used
1826 to extend or reduce the frequency range to be boosted or cut.
1827 The default value is @code{100} Hz.
1830 Set method to specify band-width of filter.
1843 Determine how steep is the filter's shelf transition.
1846 Specify which channels to filter, by default all available are filtered.
1851 Apply a biquad IIR filter with the given coefficients.
1852 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1853 are the numerator and denominator coefficients respectively.
1854 and @var{channels}, @var{c} specify which channels to filter, by default all
1855 available are filtered.
1858 Bauer stereo to binaural transformation, which improves headphone listening of
1859 stereo audio records.
1861 It accepts the following parameters:
1865 Pre-defined crossfeed level.
1869 Default level (fcut=700, feed=50).
1872 Chu Moy circuit (fcut=700, feed=60).
1875 Jan Meier circuit (fcut=650, feed=95).
1880 Cut frequency (in Hz).
1889 Remap input channels to new locations.
1891 It accepts the following parameters:
1894 Map channels from input to output. The argument is a '|'-separated list of
1895 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1896 @var{in_channel} form. @var{in_channel} can be either the name of the input
1897 channel (e.g. FL for front left) or its index in the input channel layout.
1898 @var{out_channel} is the name of the output channel or its index in the output
1899 channel layout. If @var{out_channel} is not given then it is implicitly an
1900 index, starting with zero and increasing by one for each mapping.
1902 @item channel_layout
1903 The channel layout of the output stream.
1906 If no mapping is present, the filter will implicitly map input channels to
1907 output channels, preserving indices.
1909 For example, assuming a 5.1+downmix input MOV file,
1911 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1913 will create an output WAV file tagged as stereo from the downmix channels of
1916 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1918 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1921 @section channelsplit
1923 Split each channel from an input audio stream into a separate output stream.
1925 It accepts the following parameters:
1927 @item channel_layout
1928 The channel layout of the input stream. The default is "stereo".
1931 For example, assuming a stereo input MP3 file,
1933 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1935 will create an output Matroska file with two audio streams, one containing only
1936 the left channel and the other the right channel.
1938 Split a 5.1 WAV file into per-channel files:
1940 ffmpeg -i in.wav -filter_complex
1941 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1942 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1943 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1948 Add a chorus effect to the audio.
1950 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1952 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1953 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1954 The modulation depth defines the range the modulated delay is played before or after
1955 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1956 sound tuned around the original one, like in a chorus where some vocals are slightly
1959 It accepts the following parameters:
1962 Set input gain. Default is 0.4.
1965 Set output gain. Default is 0.4.
1968 Set delays. A typical delay is around 40ms to 60ms.
1980 @subsection Examples
1986 chorus=0.7:0.9:55:0.4:0.25:2
1992 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1996 Fuller sounding chorus with three delays:
1998 chorus=0.5:0.9:50|60|40:0.4|0.32|0.3:0.25|0.4|0.3:2|2.3|1.3
2003 Compress or expand the audio's dynamic range.
2005 It accepts the following parameters:
2011 A list of times in seconds for each channel over which the instantaneous level
2012 of the input signal is averaged to determine its volume. @var{attacks} refers to
2013 increase of volume and @var{decays} refers to decrease of volume. For most
2014 situations, the attack time (response to the audio getting louder) should be
2015 shorter than the decay time, because the human ear is more sensitive to sudden
2016 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
2017 a typical value for decay is 0.8 seconds.
2018 If specified number of attacks & decays is lower than number of channels, the last
2019 set attack/decay will be used for all remaining channels.
2022 A list of points for the transfer function, specified in dB relative to the
2023 maximum possible signal amplitude. Each key points list must be defined using
2024 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
2025 @code{x0/y0 x1/y1 x2/y2 ....}
2027 The input values must be in strictly increasing order but the transfer function
2028 does not have to be monotonically rising. The point @code{0/0} is assumed but
2029 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
2030 function are @code{-70/-70|-60/-20}.
2033 Set the curve radius in dB for all joints. It defaults to 0.01.
2036 Set the additional gain in dB to be applied at all points on the transfer
2037 function. This allows for easy adjustment of the overall gain.
2041 Set an initial volume, in dB, to be assumed for each channel when filtering
2042 starts. This permits the user to supply a nominal level initially, so that, for
2043 example, a very large gain is not applied to initial signal levels before the
2044 companding has begun to operate. A typical value for audio which is initially
2045 quiet is -90 dB. It defaults to 0.
2048 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
2049 delayed before being fed to the volume adjuster. Specifying a delay
2050 approximately equal to the attack/decay times allows the filter to effectively
2051 operate in predictive rather than reactive mode. It defaults to 0.
2055 @subsection Examples
2059 Make music with both quiet and loud passages suitable for listening to in a
2062 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2065 Another example for audio with whisper and explosion parts:
2067 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2071 A noise gate for when the noise is at a lower level than the signal:
2073 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2077 Here is another noise gate, this time for when the noise is at a higher level
2078 than the signal (making it, in some ways, similar to squelch):
2080 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2084 2:1 compression starting at -6dB:
2086 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2090 2:1 compression starting at -9dB:
2092 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2096 2:1 compression starting at -12dB:
2098 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2102 2:1 compression starting at -18dB:
2104 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2108 3:1 compression starting at -15dB:
2110 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2116 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2122 compand=attacks=0:points=-80/-169|-54/-80|-49.5/-64.6|-41.1/-41.1|-25.8/-15|-10.8/-4.5|0/0|20/8.3
2126 Hard limiter at -6dB:
2128 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2132 Hard limiter at -12dB:
2134 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2138 Hard noise gate at -35 dB:
2140 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2146 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2150 @section compensationdelay
2152 Compensation Delay Line is a metric based delay to compensate differing
2153 positions of microphones or speakers.
2155 For example, you have recorded guitar with two microphones placed in
2156 different location. Because the front of sound wave has fixed speed in
2157 normal conditions, the phasing of microphones can vary and depends on
2158 their location and interposition. The best sound mix can be achieved when
2159 these microphones are in phase (synchronized). Note that distance of
2160 ~30 cm between microphones makes one microphone to capture signal in
2161 antiphase to another microphone. That makes the final mix sounding moody.
2162 This filter helps to solve phasing problems by adding different delays
2163 to each microphone track and make them synchronized.
2165 The best result can be reached when you take one track as base and
2166 synchronize other tracks one by one with it.
2167 Remember that synchronization/delay tolerance depends on sample rate, too.
2168 Higher sample rates will give more tolerance.
2170 It accepts the following parameters:
2174 Set millimeters distance. This is compensation distance for fine tuning.
2178 Set cm distance. This is compensation distance for tightening distance setup.
2182 Set meters distance. This is compensation distance for hard distance setup.
2186 Set dry amount. Amount of unprocessed (dry) signal.
2190 Set wet amount. Amount of processed (wet) signal.
2194 Set temperature degree in Celsius. This is the temperature of the environment.
2198 @section crystalizer
2199 Simple algorithm to expand audio dynamic range.
2201 The filter accepts the following options:
2205 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2206 (unchanged sound) to 10.0 (maximum effect).
2209 Enable clipping. By default is enabled.
2213 Apply a DC shift to the audio.
2215 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2216 in the recording chain) from the audio. The effect of a DC offset is reduced
2217 headroom and hence volume. The @ref{astats} filter can be used to determine if
2218 a signal has a DC offset.
2222 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2226 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2227 used to prevent clipping.
2231 Dynamic Audio Normalizer.
2233 This filter applies a certain amount of gain to the input audio in order
2234 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2235 contrast to more "simple" normalization algorithms, the Dynamic Audio
2236 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2237 This allows for applying extra gain to the "quiet" sections of the audio
2238 while avoiding distortions or clipping the "loud" sections. In other words:
2239 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2240 sections, in the sense that the volume of each section is brought to the
2241 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2242 this goal *without* applying "dynamic range compressing". It will retain 100%
2243 of the dynamic range *within* each section of the audio file.
2247 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2248 Default is 500 milliseconds.
2249 The Dynamic Audio Normalizer processes the input audio in small chunks,
2250 referred to as frames. This is required, because a peak magnitude has no
2251 meaning for just a single sample value. Instead, we need to determine the
2252 peak magnitude for a contiguous sequence of sample values. While a "standard"
2253 normalizer would simply use the peak magnitude of the complete file, the
2254 Dynamic Audio Normalizer determines the peak magnitude individually for each
2255 frame. The length of a frame is specified in milliseconds. By default, the
2256 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2257 been found to give good results with most files.
2258 Note that the exact frame length, in number of samples, will be determined
2259 automatically, based on the sampling rate of the individual input audio file.
2262 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2263 number. Default is 31.
2264 Probably the most important parameter of the Dynamic Audio Normalizer is the
2265 @code{window size} of the Gaussian smoothing filter. The filter's window size
2266 is specified in frames, centered around the current frame. For the sake of
2267 simplicity, this must be an odd number. Consequently, the default value of 31
2268 takes into account the current frame, as well as the 15 preceding frames and
2269 the 15 subsequent frames. Using a larger window results in a stronger
2270 smoothing effect and thus in less gain variation, i.e. slower gain
2271 adaptation. Conversely, using a smaller window results in a weaker smoothing
2272 effect and thus in more gain variation, i.e. faster gain adaptation.
2273 In other words, the more you increase this value, the more the Dynamic Audio
2274 Normalizer will behave like a "traditional" normalization filter. On the
2275 contrary, the more you decrease this value, the more the Dynamic Audio
2276 Normalizer will behave like a dynamic range compressor.
2279 Set the target peak value. This specifies the highest permissible magnitude
2280 level for the normalized audio input. This filter will try to approach the
2281 target peak magnitude as closely as possible, but at the same time it also
2282 makes sure that the normalized signal will never exceed the peak magnitude.
2283 A frame's maximum local gain factor is imposed directly by the target peak
2284 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2285 It is not recommended to go above this value.
2288 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2289 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2290 factor for each input frame, i.e. the maximum gain factor that does not
2291 result in clipping or distortion. The maximum gain factor is determined by
2292 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2293 additionally bounds the frame's maximum gain factor by a predetermined
2294 (global) maximum gain factor. This is done in order to avoid excessive gain
2295 factors in "silent" or almost silent frames. By default, the maximum gain
2296 factor is 10.0, For most inputs the default value should be sufficient and
2297 it usually is not recommended to increase this value. Though, for input
2298 with an extremely low overall volume level, it may be necessary to allow even
2299 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2300 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2301 Instead, a "sigmoid" threshold function will be applied. This way, the
2302 gain factors will smoothly approach the threshold value, but never exceed that
2306 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2307 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2308 This means that the maximum local gain factor for each frame is defined
2309 (only) by the frame's highest magnitude sample. This way, the samples can
2310 be amplified as much as possible without exceeding the maximum signal
2311 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2312 Normalizer can also take into account the frame's root mean square,
2313 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2314 determine the power of a time-varying signal. It is therefore considered
2315 that the RMS is a better approximation of the "perceived loudness" than
2316 just looking at the signal's peak magnitude. Consequently, by adjusting all
2317 frames to a constant RMS value, a uniform "perceived loudness" can be
2318 established. If a target RMS value has been specified, a frame's local gain
2319 factor is defined as the factor that would result in exactly that RMS value.
2320 Note, however, that the maximum local gain factor is still restricted by the
2321 frame's highest magnitude sample, in order to prevent clipping.
2324 Enable channels coupling. By default is enabled.
2325 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2326 amount. This means the same gain factor will be applied to all channels, i.e.
2327 the maximum possible gain factor is determined by the "loudest" channel.
2328 However, in some recordings, it may happen that the volume of the different
2329 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2330 In this case, this option can be used to disable the channel coupling. This way,
2331 the gain factor will be determined independently for each channel, depending
2332 only on the individual channel's highest magnitude sample. This allows for
2333 harmonizing the volume of the different channels.
2336 Enable DC bias correction. By default is disabled.
2337 An audio signal (in the time domain) is a sequence of sample values.
2338 In the Dynamic Audio Normalizer these sample values are represented in the
2339 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2340 audio signal, or "waveform", should be centered around the zero point.
2341 That means if we calculate the mean value of all samples in a file, or in a
2342 single frame, then the result should be 0.0 or at least very close to that
2343 value. If, however, there is a significant deviation of the mean value from
2344 0.0, in either positive or negative direction, this is referred to as a
2345 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2346 Audio Normalizer provides optional DC bias correction.
2347 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2348 the mean value, or "DC correction" offset, of each input frame and subtract
2349 that value from all of the frame's sample values which ensures those samples
2350 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2351 boundaries, the DC correction offset values will be interpolated smoothly
2352 between neighbouring frames.
2355 Enable alternative boundary mode. By default is disabled.
2356 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2357 around each frame. This includes the preceding frames as well as the
2358 subsequent frames. However, for the "boundary" frames, located at the very
2359 beginning and at the very end of the audio file, not all neighbouring
2360 frames are available. In particular, for the first few frames in the audio
2361 file, the preceding frames are not known. And, similarly, for the last few
2362 frames in the audio file, the subsequent frames are not known. Thus, the
2363 question arises which gain factors should be assumed for the missing frames
2364 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2365 to deal with this situation. The default boundary mode assumes a gain factor
2366 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2367 "fade out" at the beginning and at the end of the input, respectively.
2370 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2371 By default, the Dynamic Audio Normalizer does not apply "traditional"
2372 compression. This means that signal peaks will not be pruned and thus the
2373 full dynamic range will be retained within each local neighbourhood. However,
2374 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2375 normalization algorithm with a more "traditional" compression.
2376 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2377 (thresholding) function. If (and only if) the compression feature is enabled,
2378 all input frames will be processed by a soft knee thresholding function prior
2379 to the actual normalization process. Put simply, the thresholding function is
2380 going to prune all samples whose magnitude exceeds a certain threshold value.
2381 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2382 value. Instead, the threshold value will be adjusted for each individual
2384 In general, smaller parameters result in stronger compression, and vice versa.
2385 Values below 3.0 are not recommended, because audible distortion may appear.
2390 Make audio easier to listen to on headphones.
2392 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2393 so that when listened to on headphones the stereo image is moved from
2394 inside your head (standard for headphones) to outside and in front of
2395 the listener (standard for speakers).
2401 Apply a two-pole peaking equalisation (EQ) filter. With this
2402 filter, the signal-level at and around a selected frequency can
2403 be increased or decreased, whilst (unlike bandpass and bandreject
2404 filters) that at all other frequencies is unchanged.
2406 In order to produce complex equalisation curves, this filter can
2407 be given several times, each with a different central frequency.
2409 The filter accepts the following options:
2413 Set the filter's central frequency in Hz.
2416 Set method to specify band-width of filter.
2429 Specify the band-width of a filter in width_type units.
2432 Set the required gain or attenuation in dB.
2433 Beware of clipping when using a positive gain.
2436 Specify which channels to filter, by default all available are filtered.
2439 @subsection Examples
2442 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2444 equalizer=f=1000:width_type=h:width=200:g=-10
2448 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2450 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2454 @section extrastereo
2456 Linearly increases the difference between left and right channels which
2457 adds some sort of "live" effect to playback.
2459 The filter accepts the following options:
2463 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2464 (average of both channels), with 1.0 sound will be unchanged, with
2465 -1.0 left and right channels will be swapped.
2468 Enable clipping. By default is enabled.
2471 @section firequalizer
2472 Apply FIR Equalization using arbitrary frequency response.
2474 The filter accepts the following option:
2478 Set gain curve equation (in dB). The expression can contain variables:
2481 the evaluated frequency
2485 channel number, set to 0 when multichannels evaluation is disabled
2487 channel id, see libavutil/channel_layout.h, set to the first channel id when
2488 multichannels evaluation is disabled
2492 channel_layout, see libavutil/channel_layout.h
2497 @item gain_interpolate(f)
2498 interpolate gain on frequency f based on gain_entry
2499 @item cubic_interpolate(f)
2500 same as gain_interpolate, but smoother
2502 This option is also available as command. Default is @code{gain_interpolate(f)}.
2505 Set gain entry for gain_interpolate function. The expression can
2509 store gain entry at frequency f with value g
2511 This option is also available as command.
2514 Set filter delay in seconds. Higher value means more accurate.
2515 Default is @code{0.01}.
2518 Set filter accuracy in Hz. Lower value means more accurate.
2519 Default is @code{5}.
2522 Set window function. Acceptable values are:
2525 rectangular window, useful when gain curve is already smooth
2527 hann window (default)
2533 3-terms continuous 1st derivative nuttall window
2535 minimum 3-terms discontinuous nuttall window
2537 4-terms continuous 1st derivative nuttall window
2539 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2541 blackman-harris window
2547 If enabled, use fixed number of audio samples. This improves speed when
2548 filtering with large delay. Default is disabled.
2551 Enable multichannels evaluation on gain. Default is disabled.
2554 Enable zero phase mode by subtracting timestamp to compensate delay.
2555 Default is disabled.
2558 Set scale used by gain. Acceptable values are:
2561 linear frequency, linear gain
2563 linear frequency, logarithmic (in dB) gain (default)
2565 logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2567 logarithmic frequency, logarithmic gain
2571 Set file for dumping, suitable for gnuplot.
2574 Set scale for dumpfile. Acceptable values are same with scale option.
2578 Enable 2-channel convolution using complex FFT. This improves speed significantly.
2579 Default is disabled.
2582 @subsection Examples
2587 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2590 lowpass at 1000 Hz with gain_entry:
2592 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2595 custom equalization:
2597 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2600 higher delay with zero phase to compensate delay:
2602 firequalizer=delay=0.1:fixed=on:zero_phase=on
2605 lowpass on left channel, highpass on right channel:
2607 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2608 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2613 Apply a flanging effect to the audio.
2615 The filter accepts the following options:
2619 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2622 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2625 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2629 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2630 Default value is 71.
2633 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2636 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2637 Default value is @var{sinusoidal}.
2640 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2641 Default value is 25.
2644 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2645 Default is @var{linear}.
2650 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
2651 embedded HDCD codes is expanded into a 20-bit PCM stream.
2653 The filter supports the Peak Extend and Low-level Gain Adjustment features
2654 of HDCD, and detects the Transient Filter flag.
2657 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
2660 When using the filter with wav, note the default encoding for wav is 16-bit,
2661 so the resulting 20-bit stream will be truncated back to 16-bit. Use something
2662 like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
2664 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2665 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2668 The filter accepts the following options:
2671 @item disable_autoconvert
2672 Disable any automatic format conversion or resampling in the filter graph.
2674 @item process_stereo
2675 Process the stereo channels together. If target_gain does not match between
2676 channels, consider it invalid and use the last valid target_gain.
2679 Set the code detect timer period in ms.
2682 Always extend peaks above -3dBFS even if PE isn't signaled.
2685 Replace audio with a solid tone and adjust the amplitude to signal some
2686 specific aspect of the decoding process. The output file can be loaded in
2687 an audio editor alongside the original to aid analysis.
2689 @code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2696 Gain adjustment level at each sample
2698 Samples where peak extend occurs
2700 Samples where the code detect timer is active
2702 Samples where the target gain does not match between channels
2708 Apply a high-pass filter with 3dB point frequency.
2709 The filter can be either single-pole, or double-pole (the default).
2710 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2712 The filter accepts the following options:
2716 Set frequency in Hz. Default is 3000.
2719 Set number of poles. Default is 2.
2722 Set method to specify band-width of filter.
2735 Specify the band-width of a filter in width_type units.
2736 Applies only to double-pole filter.
2737 The default is 0.707q and gives a Butterworth response.
2740 Specify which channels to filter, by default all available are filtered.
2745 Join multiple input streams into one multi-channel stream.
2747 It accepts the following parameters:
2751 The number of input streams. It defaults to 2.
2753 @item channel_layout
2754 The desired output channel layout. It defaults to stereo.
2757 Map channels from inputs to output. The argument is a '|'-separated list of
2758 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2759 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2760 can be either the name of the input channel (e.g. FL for front left) or its
2761 index in the specified input stream. @var{out_channel} is the name of the output
2765 The filter will attempt to guess the mappings when they are not specified
2766 explicitly. It does so by first trying to find an unused matching input channel
2767 and if that fails it picks the first unused input channel.
2769 Join 3 inputs (with properly set channel layouts):
2771 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2774 Build a 5.1 output from 6 single-channel streams:
2776 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2777 'join=inputs=6:channel_layout=5.1:map=0.0-FL|1.0-FR|2.0-FC|3.0-SL|4.0-SR|5.0-LFE'
2783 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2785 To enable compilation of this filter you need to configure FFmpeg with
2786 @code{--enable-ladspa}.
2790 Specifies the name of LADSPA plugin library to load. If the environment
2791 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2792 each one of the directories specified by the colon separated list in
2793 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2794 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2795 @file{/usr/lib/ladspa/}.
2798 Specifies the plugin within the library. Some libraries contain only
2799 one plugin, but others contain many of them. If this is not set filter
2800 will list all available plugins within the specified library.
2803 Set the '|' separated list of controls which are zero or more floating point
2804 values that determine the behavior of the loaded plugin (for example delay,
2806 Controls need to be defined using the following syntax:
2807 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2808 @var{valuei} is the value set on the @var{i}-th control.
2809 Alternatively they can be also defined using the following syntax:
2810 @var{value0}|@var{value1}|@var{value2}|..., where
2811 @var{valuei} is the value set on the @var{i}-th control.
2812 If @option{controls} is set to @code{help}, all available controls and
2813 their valid ranges are printed.
2815 @item sample_rate, s
2816 Specify the sample rate, default to 44100. Only used if plugin have
2820 Set the number of samples per channel per each output frame, default
2821 is 1024. Only used if plugin have zero inputs.
2824 Set the minimum duration of the sourced audio. See
2825 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2826 for the accepted syntax.
2827 Note that the resulting duration may be greater than the specified duration,
2828 as the generated audio is always cut at the end of a complete frame.
2829 If not specified, or the expressed duration is negative, the audio is
2830 supposed to be generated forever.
2831 Only used if plugin have zero inputs.
2835 @subsection Examples
2839 List all available plugins within amp (LADSPA example plugin) library:
2845 List all available controls and their valid ranges for @code{vcf_notch}
2846 plugin from @code{VCF} library:
2848 ladspa=f=vcf:p=vcf_notch:c=help
2852 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2855 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2859 Add reverberation to the audio using TAP-plugins
2860 (Tom's Audio Processing plugins):
2862 ladspa=file=tap_reverb:tap_reverb
2866 Generate white noise, with 0.2 amplitude:
2868 ladspa=file=cmt:noise_source_white:c=c0=.2
2872 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2873 @code{C* Audio Plugin Suite} (CAPS) library:
2875 ladspa=file=caps:Click:c=c1=20'
2879 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2881 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2885 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2886 @code{SWH Plugins} collection:
2888 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2892 Attenuate low frequencies using Multiband EQ from Steve Harris
2893 @code{SWH Plugins} collection:
2895 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2899 @subsection Commands
2901 This filter supports the following commands:
2904 Modify the @var{N}-th control value.
2906 If the specified value is not valid, it is ignored and prior one is kept.
2911 EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2912 Support for both single pass (livestreams, files) and double pass (files) modes.
2913 This algorithm can target IL, LRA, and maximum true peak.
2915 The filter accepts the following options:
2919 Set integrated loudness target.
2920 Range is -70.0 - -5.0. Default value is -24.0.
2923 Set loudness range target.
2924 Range is 1.0 - 20.0. Default value is 7.0.
2927 Set maximum true peak.
2928 Range is -9.0 - +0.0. Default value is -2.0.
2930 @item measured_I, measured_i
2931 Measured IL of input file.
2932 Range is -99.0 - +0.0.
2934 @item measured_LRA, measured_lra
2935 Measured LRA of input file.
2936 Range is 0.0 - 99.0.
2938 @item measured_TP, measured_tp
2939 Measured true peak of input file.
2940 Range is -99.0 - +99.0.
2942 @item measured_thresh
2943 Measured threshold of input file.
2944 Range is -99.0 - +0.0.
2947 Set offset gain. Gain is applied before the true-peak limiter.
2948 Range is -99.0 - +99.0. Default is +0.0.
2951 Normalize linearly if possible.
2952 measured_I, measured_LRA, measured_TP, and measured_thresh must also
2953 to be specified in order to use this mode.
2954 Options are true or false. Default is true.
2957 Treat mono input files as "dual-mono". If a mono file is intended for playback
2958 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
2959 If set to @code{true}, this option will compensate for this effect.
2960 Multi-channel input files are not affected by this option.
2961 Options are true or false. Default is false.
2964 Set print format for stats. Options are summary, json, or none.
2965 Default value is none.
2970 Apply a low-pass filter with 3dB point frequency.
2971 The filter can be either single-pole or double-pole (the default).
2972 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2974 The filter accepts the following options:
2978 Set frequency in Hz. Default is 500.
2981 Set number of poles. Default is 2.
2984 Set method to specify band-width of filter.
2997 Specify the band-width of a filter in width_type units.
2998 Applies only to double-pole filter.
2999 The default is 0.707q and gives a Butterworth response.
3002 Specify which channels to filter, by default all available are filtered.
3005 @subsection Examples
3008 Lowpass only LFE channel, it LFE is not present it does nothing:
3017 Mix channels with specific gain levels. The filter accepts the output
3018 channel layout followed by a set of channels definitions.
3020 This filter is also designed to efficiently remap the channels of an audio
3023 The filter accepts parameters of the form:
3024 "@var{l}|@var{outdef}|@var{outdef}|..."
3028 output channel layout or number of channels
3031 output channel specification, of the form:
3032 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3035 output channel to define, either a channel name (FL, FR, etc.) or a channel
3036 number (c0, c1, etc.)
3039 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3042 input channel to use, see out_name for details; it is not possible to mix
3043 named and numbered input channels
3046 If the `=' in a channel specification is replaced by `<', then the gains for
3047 that specification will be renormalized so that the total is 1, thus
3048 avoiding clipping noise.
3050 @subsection Mixing examples
3052 For example, if you want to down-mix from stereo to mono, but with a bigger
3053 factor for the left channel:
3055 pan=1c|c0=0.9*c0+0.1*c1
3058 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3059 7-channels surround:
3061 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3064 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3065 that should be preferred (see "-ac" option) unless you have very specific
3068 @subsection Remapping examples
3070 The channel remapping will be effective if, and only if:
3073 @item gain coefficients are zeroes or ones,
3074 @item only one input per channel output,
3077 If all these conditions are satisfied, the filter will notify the user ("Pure
3078 channel mapping detected"), and use an optimized and lossless method to do the
3081 For example, if you have a 5.1 source and want a stereo audio stream by
3082 dropping the extra channels:
3084 pan="stereo| c0=FL | c1=FR"
3087 Given the same source, you can also switch front left and front right channels
3088 and keep the input channel layout:
3090 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3093 If the input is a stereo audio stream, you can mute the front left channel (and
3094 still keep the stereo channel layout) with:
3099 Still with a stereo audio stream input, you can copy the right channel in both
3100 front left and right:
3102 pan="stereo| c0=FR | c1=FR"
3107 ReplayGain scanner filter. This filter takes an audio stream as an input and
3108 outputs it unchanged.
3109 At end of filtering it displays @code{track_gain} and @code{track_peak}.
3113 Convert the audio sample format, sample rate and channel layout. It is
3114 not meant to be used directly.
3117 Apply time-stretching and pitch-shifting with librubberband.
3119 The filter accepts the following options:
3123 Set tempo scale factor.
3126 Set pitch scale factor.
3129 Set transients detector.
3130 Possible values are:
3139 Possible values are:
3148 Possible values are:
3155 Set processing window size.
3156 Possible values are:
3165 Possible values are:
3172 Enable formant preservation when shift pitching.
3173 Possible values are:
3181 Possible values are:
3190 Possible values are:
3197 @section sidechaincompress
3199 This filter acts like normal compressor but has the ability to compress
3200 detected signal using second input signal.
3201 It needs two input streams and returns one output stream.
3202 First input stream will be processed depending on second stream signal.
3203 The filtered signal then can be filtered with other filters in later stages of
3204 processing. See @ref{pan} and @ref{amerge} filter.
3206 The filter accepts the following options:
3210 Set input gain. Default is 1. Range is between 0.015625 and 64.
3213 If a signal of second stream raises above this level it will affect the gain
3214 reduction of first stream.
3215 By default is 0.125. Range is between 0.00097563 and 1.
3218 Set a ratio about which the signal is reduced. 1:2 means that if the level
3219 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3220 Default is 2. Range is between 1 and 20.
3223 Amount of milliseconds the signal has to rise above the threshold before gain
3224 reduction starts. Default is 20. Range is between 0.01 and 2000.
3227 Amount of milliseconds the signal has to fall below the threshold before
3228 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3231 Set the amount by how much signal will be amplified after processing.
3232 Default is 2. Range is from 1 and 64.
3235 Curve the sharp knee around the threshold to enter gain reduction more softly.
3236 Default is 2.82843. Range is between 1 and 8.
3239 Choose if the @code{average} level between all channels of side-chain stream
3240 or the louder(@code{maximum}) channel of side-chain stream affects the
3241 reduction. Default is @code{average}.
3244 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3245 of @code{rms}. Default is @code{rms} which is mainly smoother.
3248 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3251 How much to use compressed signal in output. Default is 1.
3252 Range is between 0 and 1.
3255 @subsection Examples
3259 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3260 depending on the signal of 2nd input and later compressed signal to be
3261 merged with 2nd input:
3263 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3267 @section sidechaingate
3269 A sidechain gate acts like a normal (wideband) gate but has the ability to
3270 filter the detected signal before sending it to the gain reduction stage.
3271 Normally a gate uses the full range signal to detect a level above the
3273 For example: If you cut all lower frequencies from your sidechain signal
3274 the gate will decrease the volume of your track only if not enough highs
3275 appear. With this technique you are able to reduce the resonation of a
3276 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3278 It needs two input streams and returns one output stream.
3279 First input stream will be processed depending on second stream signal.
3281 The filter accepts the following options:
3285 Set input level before filtering.
3286 Default is 1. Allowed range is from 0.015625 to 64.
3289 Set the level of gain reduction when the signal is below the threshold.
3290 Default is 0.06125. Allowed range is from 0 to 1.
3293 If a signal rises above this level the gain reduction is released.
3294 Default is 0.125. Allowed range is from 0 to 1.
3297 Set a ratio about which the signal is reduced.
3298 Default is 2. Allowed range is from 1 to 9000.
3301 Amount of milliseconds the signal has to rise above the threshold before gain
3303 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3306 Amount of milliseconds the signal has to fall below the threshold before the
3307 reduction is increased again. Default is 250 milliseconds.
3308 Allowed range is from 0.01 to 9000.
3311 Set amount of amplification of signal after processing.
3312 Default is 1. Allowed range is from 1 to 64.
3315 Curve the sharp knee around the threshold to enter gain reduction more softly.
3316 Default is 2.828427125. Allowed range is from 1 to 8.
3319 Choose if exact signal should be taken for detection or an RMS like one.
3320 Default is rms. Can be peak or rms.
3323 Choose if the average level between all channels or the louder channel affects
3325 Default is average. Can be average or maximum.
3328 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3331 @section silencedetect
3333 Detect silence in an audio stream.
3335 This filter logs a message when it detects that the input audio volume is less
3336 or equal to a noise tolerance value for a duration greater or equal to the
3337 minimum detected noise duration.
3339 The printed times and duration are expressed in seconds.
3341 The filter accepts the following options:
3345 Set silence duration until notification (default is 2 seconds).
3348 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3349 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3352 @subsection Examples
3356 Detect 5 seconds of silence with -50dB noise tolerance:
3358 silencedetect=n=-50dB:d=5
3362 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3363 tolerance in @file{silence.mp3}:
3365 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3369 @section silenceremove
3371 Remove silence from the beginning, middle or end of the audio.
3373 The filter accepts the following options:
3377 This value is used to indicate if audio should be trimmed at beginning of
3378 the audio. A value of zero indicates no silence should be trimmed from the
3379 beginning. When specifying a non-zero value, it trims audio up until it
3380 finds non-silence. Normally, when trimming silence from beginning of audio
3381 the @var{start_periods} will be @code{1} but it can be increased to higher
3382 values to trim all audio up to specific count of non-silence periods.
3383 Default value is @code{0}.
3385 @item start_duration
3386 Specify the amount of time that non-silence must be detected before it stops
3387 trimming audio. By increasing the duration, bursts of noises can be treated
3388 as silence and trimmed off. Default value is @code{0}.
3390 @item start_threshold
3391 This indicates what sample value should be treated as silence. For digital
3392 audio, a value of @code{0} may be fine but for audio recorded from analog,
3393 you may wish to increase the value to account for background noise.
3394 Can be specified in dB (in case "dB" is appended to the specified value)
3395 or amplitude ratio. Default value is @code{0}.
3398 Set the count for trimming silence from the end of audio.
3399 To remove silence from the middle of a file, specify a @var{stop_periods}
3400 that is negative. This value is then treated as a positive value and is
3401 used to indicate the effect should restart processing as specified by
3402 @var{start_periods}, making it suitable for removing periods of silence
3403 in the middle of the audio.
3404 Default value is @code{0}.
3407 Specify a duration of silence that must exist before audio is not copied any
3408 more. By specifying a higher duration, silence that is wanted can be left in
3410 Default value is @code{0}.
3412 @item stop_threshold
3413 This is the same as @option{start_threshold} but for trimming silence from
3415 Can be specified in dB (in case "dB" is appended to the specified value)
3416 or amplitude ratio. Default value is @code{0}.
3419 This indicates that @var{stop_duration} length of audio should be left intact
3420 at the beginning of each period of silence.
3421 For example, if you want to remove long pauses between words but do not want
3422 to remove the pauses completely. Default value is @code{0}.
3425 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3426 and works better with digital silence which is exactly 0.
3427 Default value is @code{rms}.
3430 Set ratio used to calculate size of window for detecting silence.
3431 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3434 @subsection Examples
3438 The following example shows how this filter can be used to start a recording
3439 that does not contain the delay at the start which usually occurs between
3440 pressing the record button and the start of the performance:
3442 silenceremove=1:5:0.02
3446 Trim all silence encountered from beginning to end where there is more than 1
3447 second of silence in audio:
3449 silenceremove=0:0:0:-1:1:-90dB
3455 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3456 loudspeakers around the user for binaural listening via headphones (audio
3457 formats up to 9 channels supported).
3458 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3459 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3460 Austrian Academy of Sciences.
3462 To enable compilation of this filter you need to configure FFmpeg with
3463 @code{--enable-netcdf}.
3465 The filter accepts the following options:
3469 Set the SOFA file used for rendering.
3472 Set gain applied to audio. Value is in dB. Default is 0.
3475 Set rotation of virtual loudspeakers in deg. Default is 0.
3478 Set elevation of virtual speakers in deg. Default is 0.
3481 Set distance in meters between loudspeakers and the listener with near-field
3482 HRTFs. Default is 1.
3485 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3486 processing audio in time domain which is slow.
3487 @var{freq} is processing audio in frequency domain which is fast.
3488 Default is @var{freq}.
3491 Set custom positions of virtual loudspeakers. Syntax for this option is:
3492 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3493 Each virtual loudspeaker is described with short channel name following with
3494 azimuth and elevation in degreees.
3495 Each virtual loudspeaker description is separated by '|'.
3496 For example to override front left and front right channel positions use:
3497 'speakers=FL 45 15|FR 345 15'.
3498 Descriptions with unrecognised channel names are ignored.
3501 @subsection Examples
3505 Using ClubFritz6 sofa file:
3507 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3511 Using ClubFritz12 sofa file and bigger radius with small rotation:
3513 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3517 Similar as above but with custom speaker positions for front left, front right, back left and back right
3518 and also with custom gain:
3520 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3524 @section stereotools
3526 This filter has some handy utilities to manage stereo signals, for converting
3527 M/S stereo recordings to L/R signal while having control over the parameters
3528 or spreading the stereo image of master track.
3530 The filter accepts the following options:
3534 Set input level before filtering for both channels. Defaults is 1.
3535 Allowed range is from 0.015625 to 64.
3538 Set output level after filtering for both channels. Defaults is 1.
3539 Allowed range is from 0.015625 to 64.
3542 Set input balance between both channels. Default is 0.
3543 Allowed range is from -1 to 1.
3546 Set output balance between both channels. Default is 0.
3547 Allowed range is from -1 to 1.
3550 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3551 clipping. Disabled by default.
3554 Mute the left channel. Disabled by default.
3557 Mute the right channel. Disabled by default.
3560 Change the phase of the left channel. Disabled by default.
3563 Change the phase of the right channel. Disabled by default.
3566 Set stereo mode. Available values are:
3570 Left/Right to Left/Right, this is default.
3573 Left/Right to Mid/Side.
3576 Mid/Side to Left/Right.
3579 Left/Right to Left/Left.
3582 Left/Right to Right/Right.
3585 Left/Right to Left + Right.
3588 Left/Right to Right/Left.
3592 Set level of side signal. Default is 1.
3593 Allowed range is from 0.015625 to 64.
3596 Set balance of side signal. Default is 0.
3597 Allowed range is from -1 to 1.
3600 Set level of the middle signal. Default is 1.
3601 Allowed range is from 0.015625 to 64.
3604 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3607 Set stereo base between mono and inversed channels. Default is 0.
3608 Allowed range is from -1 to 1.
3611 Set delay in milliseconds how much to delay left from right channel and
3612 vice versa. Default is 0. Allowed range is from -20 to 20.
3615 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3618 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3621 @subsection Examples
3625 Apply karaoke like effect:
3627 stereotools=mlev=0.015625
3631 Convert M/S signal to L/R:
3633 "stereotools=mode=ms>lr"
3637 @section stereowiden
3639 This filter enhance the stereo effect by suppressing signal common to both
3640 channels and by delaying the signal of left into right and vice versa,
3641 thereby widening the stereo effect.
3643 The filter accepts the following options:
3647 Time in milliseconds of the delay of left signal into right and vice versa.
3648 Default is 20 milliseconds.
3651 Amount of gain in delayed signal into right and vice versa. Gives a delay
3652 effect of left signal in right output and vice versa which gives widening
3653 effect. Default is 0.3.
3656 Cross feed of left into right with inverted phase. This helps in suppressing
3657 the mono. If the value is 1 it will cancel all the signal common to both
3658 channels. Default is 0.3.
3661 Set level of input signal of original channel. Default is 0.8.
3666 Boost or cut treble (upper) frequencies of the audio using a two-pole
3667 shelving filter with a response similar to that of a standard
3668 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3670 The filter accepts the following options:
3674 Give the gain at whichever is the lower of ~22 kHz and the
3675 Nyquist frequency. Its useful range is about -20 (for a large cut)
3676 to +20 (for a large boost). Beware of clipping when using a positive gain.
3679 Set the filter's central frequency and so can be used
3680 to extend or reduce the frequency range to be boosted or cut.
3681 The default value is @code{3000} Hz.
3684 Set method to specify band-width of filter.
3697 Determine how steep is the filter's shelf transition.
3700 Specify which channels to filter, by default all available are filtered.
3705 Sinusoidal amplitude modulation.
3707 The filter accepts the following options:
3711 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3712 (20 Hz or lower) will result in a tremolo effect.
3713 This filter may also be used as a ring modulator by specifying
3714 a modulation frequency higher than 20 Hz.
3715 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3718 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3719 Default value is 0.5.
3724 Sinusoidal phase modulation.
3726 The filter accepts the following options:
3730 Modulation frequency in Hertz.
3731 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3734 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3735 Default value is 0.5.
3740 Adjust the input audio volume.
3742 It accepts the following parameters:
3746 Set audio volume expression.
3748 Output values are clipped to the maximum value.
3750 The output audio volume is given by the relation:
3752 @var{output_volume} = @var{volume} * @var{input_volume}
3755 The default value for @var{volume} is "1.0".
3758 This parameter represents the mathematical precision.
3760 It determines which input sample formats will be allowed, which affects the
3761 precision of the volume scaling.
3765 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3767 32-bit floating-point; this limits input sample format to FLT. (default)
3769 64-bit floating-point; this limits input sample format to DBL.
3773 Choose the behaviour on encountering ReplayGain side data in input frames.
3777 Remove ReplayGain side data, ignoring its contents (the default).
3780 Ignore ReplayGain side data, but leave it in the frame.
3783 Prefer the track gain, if present.
3786 Prefer the album gain, if present.
3789 @item replaygain_preamp
3790 Pre-amplification gain in dB to apply to the selected replaygain gain.
3792 Default value for @var{replaygain_preamp} is 0.0.
3795 Set when the volume expression is evaluated.
3797 It accepts the following values:
3800 only evaluate expression once during the filter initialization, or
3801 when the @samp{volume} command is sent
3804 evaluate expression for each incoming frame
3807 Default value is @samp{once}.
3810 The volume expression can contain the following parameters.
3814 frame number (starting at zero)
3817 @item nb_consumed_samples
3818 number of samples consumed by the filter
3820 number of samples in the current frame
3822 original frame position in the file
3828 PTS at start of stream
3830 time at start of stream
3836 last set volume value
3839 Note that when @option{eval} is set to @samp{once} only the
3840 @var{sample_rate} and @var{tb} variables are available, all other
3841 variables will evaluate to NAN.
3843 @subsection Commands
3845 This filter supports the following commands:
3848 Modify the volume expression.
3849 The command accepts the same syntax of the corresponding option.
3851 If the specified expression is not valid, it is kept at its current
3853 @item replaygain_noclip
3854 Prevent clipping by limiting the gain applied.
3856 Default value for @var{replaygain_noclip} is 1.
3860 @subsection Examples
3864 Halve the input audio volume:
3868 volume=volume=-6.0206dB
3871 In all the above example the named key for @option{volume} can be
3872 omitted, for example like in:
3878 Increase input audio power by 6 decibels using fixed-point precision:
3880 volume=volume=6dB:precision=fixed
3884 Fade volume after time 10 with an annihilation period of 5 seconds:
3886 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3890 @section volumedetect
3892 Detect the volume of the input video.
3894 The filter has no parameters. The input is not modified. Statistics about
3895 the volume will be printed in the log when the input stream end is reached.
3897 In particular it will show the mean volume (root mean square), maximum
3898 volume (on a per-sample basis), and the beginning of a histogram of the
3899 registered volume values (from the maximum value to a cumulated 1/1000 of
3902 All volumes are in decibels relative to the maximum PCM value.
3904 @subsection Examples
3906 Here is an excerpt of the output:
3908 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3909 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3910 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3911 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3912 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3913 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3914 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3915 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3916 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3922 The mean square energy is approximately -27 dB, or 10^-2.7.
3924 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3926 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3929 In other words, raising the volume by +4 dB does not cause any clipping,
3930 raising it by +5 dB causes clipping for 6 samples, etc.
3932 @c man end AUDIO FILTERS
3934 @chapter Audio Sources
3935 @c man begin AUDIO SOURCES
3937 Below is a description of the currently available audio sources.
3941 Buffer audio frames, and make them available to the filter chain.
3943 This source is mainly intended for a programmatic use, in particular
3944 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3946 It accepts the following parameters:
3950 The timebase which will be used for timestamps of submitted frames. It must be
3951 either a floating-point number or in @var{numerator}/@var{denominator} form.
3954 The sample rate of the incoming audio buffers.
3957 The sample format of the incoming audio buffers.
3958 Either a sample format name or its corresponding integer representation from
3959 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3961 @item channel_layout
3962 The channel layout of the incoming audio buffers.
3963 Either a channel layout name from channel_layout_map in
3964 @file{libavutil/channel_layout.c} or its corresponding integer representation
3965 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3968 The number of channels of the incoming audio buffers.
3969 If both @var{channels} and @var{channel_layout} are specified, then they
3974 @subsection Examples
3977 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3980 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3981 Since the sample format with name "s16p" corresponds to the number
3982 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3985 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3990 Generate an audio signal specified by an expression.
3992 This source accepts in input one or more expressions (one for each
3993 channel), which are evaluated and used to generate a corresponding
3996 This source accepts the following options:
4000 Set the '|'-separated expressions list for each separate channel. In case the
4001 @option{channel_layout} option is not specified, the selected channel layout
4002 depends on the number of provided expressions. Otherwise the last
4003 specified expression is applied to the remaining output channels.
4005 @item channel_layout, c
4006 Set the channel layout. The number of channels in the specified layout
4007 must be equal to the number of specified expressions.
4010 Set the minimum duration of the sourced audio. See
4011 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
4012 for the accepted syntax.
4013 Note that the resulting duration may be greater than the specified
4014 duration, as the generated audio is always cut at the end of a
4017 If not specified, or the expressed duration is negative, the audio is
4018 supposed to be generated forever.
4021 Set the number of samples per channel per each output frame,
4024 @item sample_rate, s
4025 Specify the sample rate, default to 44100.
4028 Each expression in @var{exprs} can contain the following constants:
4032 number of the evaluated sample, starting from 0
4035 time of the evaluated sample expressed in seconds, starting from 0
4042 @subsection Examples
4052 Generate a sin signal with frequency of 440 Hz, set sample rate to
4055 aevalsrc="sin(440*2*PI*t):s=8000"
4059 Generate a two channels signal, specify the channel layout (Front
4060 Center + Back Center) explicitly:
4062 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4066 Generate white noise:
4068 aevalsrc="-2+random(0)"
4072 Generate an amplitude modulated signal:
4074 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4078 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4080 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4087 The null audio source, return unprocessed audio frames. It is mainly useful
4088 as a template and to be employed in analysis / debugging tools, or as
4089 the source for filters which ignore the input data (for example the sox
4092 This source accepts the following options:
4096 @item channel_layout, cl
4098 Specifies the channel layout, and can be either an integer or a string
4099 representing a channel layout. The default value of @var{channel_layout}
4102 Check the channel_layout_map definition in
4103 @file{libavutil/channel_layout.c} for the mapping between strings and
4104 channel layout values.
4106 @item sample_rate, r
4107 Specifies the sample rate, and defaults to 44100.
4110 Set the number of samples per requested frames.
4114 @subsection Examples
4118 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4120 anullsrc=r=48000:cl=4
4124 Do the same operation with a more obvious syntax:
4126 anullsrc=r=48000:cl=mono
4130 All the parameters need to be explicitly defined.
4134 Synthesize a voice utterance using the libflite library.
4136 To enable compilation of this filter you need to configure FFmpeg with
4137 @code{--enable-libflite}.
4139 Note that the flite library is not thread-safe.
4141 The filter accepts the following options:
4146 If set to 1, list the names of the available voices and exit
4147 immediately. Default value is 0.
4150 Set the maximum number of samples per frame. Default value is 512.
4153 Set the filename containing the text to speak.
4156 Set the text to speak.
4159 Set the voice to use for the speech synthesis. Default value is
4160 @code{kal}. See also the @var{list_voices} option.
4163 @subsection Examples
4167 Read from file @file{speech.txt}, and synthesize the text using the
4168 standard flite voice:
4170 flite=textfile=speech.txt
4174 Read the specified text selecting the @code{slt} voice:
4176 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4180 Input text to ffmpeg:
4182 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4186 Make @file{ffplay} speak the specified text, using @code{flite} and
4187 the @code{lavfi} device:
4189 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4193 For more information about libflite, check:
4194 @url{http://www.speech.cs.cmu.edu/flite/}
4198 Generate a noise audio signal.
4200 The filter accepts the following options:
4203 @item sample_rate, r
4204 Specify the sample rate. Default value is 48000 Hz.
4207 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4211 Specify the duration of the generated audio stream. Not specifying this option
4212 results in noise with an infinite length.
4214 @item color, colour, c
4215 Specify the color of noise. Available noise colors are white, pink, and brown.
4216 Default color is white.
4219 Specify a value used to seed the PRNG.
4222 Set the number of samples per each output frame, default is 1024.
4225 @subsection Examples
4230 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4232 anoisesrc=d=60:c=pink:r=44100:a=0.5
4238 Generate an audio signal made of a sine wave with amplitude 1/8.
4240 The audio signal is bit-exact.
4242 The filter accepts the following options:
4247 Set the carrier frequency. Default is 440 Hz.
4249 @item beep_factor, b
4250 Enable a periodic beep every second with frequency @var{beep_factor} times
4251 the carrier frequency. Default is 0, meaning the beep is disabled.
4253 @item sample_rate, r
4254 Specify the sample rate, default is 44100.
4257 Specify the duration of the generated audio stream.
4259 @item samples_per_frame
4260 Set the number of samples per output frame.
4262 The expression can contain the following constants:
4266 The (sequential) number of the output audio frame, starting from 0.
4269 The PTS (Presentation TimeStamp) of the output audio frame,
4270 expressed in @var{TB} units.
4273 The PTS of the output audio frame, expressed in seconds.
4276 The timebase of the output audio frames.
4279 Default is @code{1024}.
4282 @subsection Examples
4287 Generate a simple 440 Hz sine wave:
4293 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4297 sine=frequency=220:beep_factor=4:duration=5
4301 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4304 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4308 @c man end AUDIO SOURCES
4310 @chapter Audio Sinks
4311 @c man begin AUDIO SINKS
4313 Below is a description of the currently available audio sinks.
4315 @section abuffersink
4317 Buffer audio frames, and make them available to the end of filter chain.
4319 This sink is mainly intended for programmatic use, in particular
4320 through the interface defined in @file{libavfilter/buffersink.h}
4321 or the options system.
4323 It accepts a pointer to an AVABufferSinkContext structure, which
4324 defines the incoming buffers' formats, to be passed as the opaque
4325 parameter to @code{avfilter_init_filter} for initialization.
4328 Null audio sink; do absolutely nothing with the input audio. It is
4329 mainly useful as a template and for use in analysis / debugging
4332 @c man end AUDIO SINKS
4334 @chapter Video Filters
4335 @c man begin VIDEO FILTERS
4337 When you configure your FFmpeg build, you can disable any of the
4338 existing filters using @code{--disable-filters}.
4339 The configure output will show the video filters included in your
4342 Below is a description of the currently available video filters.
4344 @section alphaextract
4346 Extract the alpha component from the input as a grayscale video. This
4347 is especially useful with the @var{alphamerge} filter.
4351 Add or replace the alpha component of the primary input with the
4352 grayscale value of a second input. This is intended for use with
4353 @var{alphaextract} to allow the transmission or storage of frame
4354 sequences that have alpha in a format that doesn't support an alpha
4357 For example, to reconstruct full frames from a normal YUV-encoded video
4358 and a separate video created with @var{alphaextract}, you might use:
4360 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4363 Since this filter is designed for reconstruction, it operates on frame
4364 sequences without considering timestamps, and terminates when either
4365 input reaches end of stream. This will cause problems if your encoding
4366 pipeline drops frames. If you're trying to apply an image as an
4367 overlay to a video stream, consider the @var{overlay} filter instead.
4371 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4372 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4373 Substation Alpha) subtitles files.
4375 This filter accepts the following option in addition to the common options from
4376 the @ref{subtitles} filter:
4380 Set the shaping engine
4382 Available values are:
4385 The default libass shaping engine, which is the best available.
4387 Fast, font-agnostic shaper that can do only substitutions
4389 Slower shaper using OpenType for substitutions and positioning
4392 The default is @code{auto}.
4396 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4398 The filter accepts the following options:
4402 Set threshold A for 1st plane. Default is 0.02.
4403 Valid range is 0 to 0.3.
4406 Set threshold B for 1st plane. Default is 0.04.
4407 Valid range is 0 to 5.
4410 Set threshold A for 2nd plane. Default is 0.02.
4411 Valid range is 0 to 0.3.
4414 Set threshold B for 2nd plane. Default is 0.04.
4415 Valid range is 0 to 5.
4418 Set threshold A for 3rd plane. Default is 0.02.
4419 Valid range is 0 to 0.3.
4422 Set threshold B for 3rd plane. Default is 0.04.
4423 Valid range is 0 to 5.
4425 Threshold A is designed to react on abrupt changes in the input signal and
4426 threshold B is designed to react on continuous changes in the input signal.
4429 Set number of frames filter will use for averaging. Default is 33. Must be odd
4430 number in range [5, 129].
4433 Set what planes of frame filter will use for averaging. Default is all.
4438 Apply average blur filter.
4440 The filter accepts the following options:
4444 Set horizontal kernel size.
4447 Set which planes to filter. By default all planes are filtered.
4450 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4451 Default is @code{0}.
4456 Compute the bounding box for the non-black pixels in the input frame
4459 This filter computes the bounding box containing all the pixels with a
4460 luminance value greater than the minimum allowed value.
4461 The parameters describing the bounding box are printed on the filter
4464 The filter accepts the following option:
4468 Set the minimal luminance value. Default is @code{16}.
4471 @section bitplanenoise
4473 Show and measure bit plane noise.
4475 The filter accepts the following options:
4479 Set which plane to analyze. Default is @code{1}.
4482 Filter out noisy pixels from @code{bitplane} set above.
4483 Default is disabled.
4486 @section blackdetect
4488 Detect video intervals that are (almost) completely black. Can be
4489 useful to detect chapter transitions, commercials, or invalid
4490 recordings. Output lines contains the time for the start, end and
4491 duration of the detected black interval expressed in seconds.
4493 In order to display the output lines, you need to set the loglevel at
4494 least to the AV_LOG_INFO value.
4496 The filter accepts the following options:
4499 @item black_min_duration, d
4500 Set the minimum detected black duration expressed in seconds. It must
4501 be a non-negative floating point number.
4503 Default value is 2.0.
4505 @item picture_black_ratio_th, pic_th
4506 Set the threshold for considering a picture "black".
4507 Express the minimum value for the ratio:
4509 @var{nb_black_pixels} / @var{nb_pixels}
4512 for which a picture is considered black.
4513 Default value is 0.98.
4515 @item pixel_black_th, pix_th
4516 Set the threshold for considering a pixel "black".
4518 The threshold expresses the maximum pixel luminance value for which a
4519 pixel is considered "black". The provided value is scaled according to
4520 the following equation:
4522 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4525 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4526 the input video format, the range is [0-255] for YUV full-range
4527 formats and [16-235] for YUV non full-range formats.
4529 Default value is 0.10.
4532 The following example sets the maximum pixel threshold to the minimum
4533 value, and detects only black intervals of 2 or more seconds:
4535 blackdetect=d=2:pix_th=0.00
4540 Detect frames that are (almost) completely black. Can be useful to
4541 detect chapter transitions or commercials. Output lines consist of
4542 the frame number of the detected frame, the percentage of blackness,
4543 the position in the file if known or -1 and the timestamp in seconds.
4545 In order to display the output lines, you need to set the loglevel at
4546 least to the AV_LOG_INFO value.
4548 This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4549 The value represents the percentage of pixels in the picture that
4550 are below the threshold value.
4552 It accepts the following parameters:
4557 The percentage of the pixels that have to be below the threshold; it defaults to
4560 @item threshold, thresh
4561 The threshold below which a pixel value is considered black; it defaults to
4566 @section blend, tblend
4568 Blend two video frames into each other.
4570 The @code{blend} filter takes two input streams and outputs one
4571 stream, the first input is the "top" layer and second input is
4572 "bottom" layer. By default, the output terminates when the longest input terminates.
4574 The @code{tblend} (time blend) filter takes two consecutive frames
4575 from one single stream, and outputs the result obtained by blending
4576 the new frame on top of the old frame.
4578 A description of the accepted options follows.
4586 Set blend mode for specific pixel component or all pixel components in case
4587 of @var{all_mode}. Default value is @code{normal}.
4589 Available values for component modes are:
4630 Set blend opacity for specific pixel component or all pixel components in case
4631 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4638 Set blend expression for specific pixel component or all pixel components in case
4639 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4641 The expressions can use the following variables:
4645 The sequential number of the filtered frame, starting from @code{0}.
4649 the coordinates of the current sample
4653 the width and height of currently filtered plane
4657 Width and height scale depending on the currently filtered plane. It is the
4658 ratio between the corresponding luma plane number of pixels and the current
4659 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4660 @code{0.5,0.5} for chroma planes.
4663 Time of the current frame, expressed in seconds.
4666 Value of pixel component at current location for first video frame (top layer).
4669 Value of pixel component at current location for second video frame (bottom layer).
4673 Force termination when the shortest input terminates. Default is
4674 @code{0}. This option is only defined for the @code{blend} filter.
4677 Continue applying the last bottom frame after the end of the stream. A value of
4678 @code{0} disable the filter after the last frame of the bottom layer is reached.
4679 Default is @code{1}. This option is only defined for the @code{blend} filter.
4682 @subsection Examples
4686 Apply transition from bottom layer to top layer in first 10 seconds:
4688 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4692 Apply 1x1 checkerboard effect:
4694 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4698 Apply uncover left effect:
4700 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4704 Apply uncover down effect:
4706 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4710 Apply uncover up-left effect:
4712 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4716 Split diagonally video and shows top and bottom layer on each side:
4718 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4722 Display differences between the current and the previous frame:
4724 tblend=all_mode=difference128
4730 Apply a boxblur algorithm to the input video.
4732 It accepts the following parameters:
4736 @item luma_radius, lr
4737 @item luma_power, lp
4738 @item chroma_radius, cr
4739 @item chroma_power, cp
4740 @item alpha_radius, ar
4741 @item alpha_power, ap
4745 A description of the accepted options follows.
4748 @item luma_radius, lr
4749 @item chroma_radius, cr
4750 @item alpha_radius, ar
4751 Set an expression for the box radius in pixels used for blurring the
4752 corresponding input plane.
4754 The radius value must be a non-negative number, and must not be
4755 greater than the value of the expression @code{min(w,h)/2} for the
4756 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4759 Default value for @option{luma_radius} is "2". If not specified,
4760 @option{chroma_radius} and @option{alpha_radius} default to the
4761 corresponding value set for @option{luma_radius}.
4763 The expressions can contain the following constants:
4767 The input width and height in pixels.
4771 The input chroma image width and height in pixels.
4775 The horizontal and vertical chroma subsample values. For example, for the
4776 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4779 @item luma_power, lp
4780 @item chroma_power, cp
4781 @item alpha_power, ap
4782 Specify how many times the boxblur filter is applied to the
4783 corresponding plane.
4785 Default value for @option{luma_power} is 2. If not specified,
4786 @option{chroma_power} and @option{alpha_power} default to the
4787 corresponding value set for @option{luma_power}.
4789 A value of 0 will disable the effect.
4792 @subsection Examples
4796 Apply a boxblur filter with the luma, chroma, and alpha radii
4799 boxblur=luma_radius=2:luma_power=1
4804 Set the luma radius to 2, and alpha and chroma radius to 0:
4806 boxblur=2:1:cr=0:ar=0
4810 Set the luma and chroma radii to a fraction of the video dimension:
4812 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4818 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4819 Deinterlacing Filter").
4821 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4822 interpolation algorithms.
4823 It accepts the following parameters:
4827 The interlacing mode to adopt. It accepts one of the following values:
4831 Output one frame for each frame.
4833 Output one frame for each field.
4836 The default value is @code{send_field}.
4839 The picture field parity assumed for the input interlaced video. It accepts one
4840 of the following values:
4844 Assume the top field is first.
4846 Assume the bottom field is first.
4848 Enable automatic detection of field parity.
4851 The default value is @code{auto}.
4852 If the interlacing is unknown or the decoder does not export this information,
4853 top field first will be assumed.
4856 Specify which frames to deinterlace. Accept one of the following
4861 Deinterlace all frames.
4863 Only deinterlace frames marked as interlaced.
4866 The default value is @code{all}.
4870 YUV colorspace color/chroma keying.
4872 The filter accepts the following options:
4876 The color which will be replaced with transparency.
4879 Similarity percentage with the key color.
4881 0.01 matches only the exact key color, while 1.0 matches everything.
4886 0.0 makes pixels either fully transparent, or not transparent at all.
4888 Higher values result in semi-transparent pixels, with a higher transparency
4889 the more similar the pixels color is to the key color.
4892 Signals that the color passed is already in YUV instead of RGB.
4894 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4895 This can be used to pass exact YUV values as hexadecimal numbers.
4898 @subsection Examples
4902 Make every green pixel in the input image transparent:
4904 ffmpeg -i input.png -vf chromakey=green out.png
4908 Overlay a greenscreen-video on top of a static black background.
4910 ffmpeg -f lavfi -i color=c=black:s=1280x720 -i video.mp4 -shortest -filter_complex "[1:v]chromakey=0x70de77:0.1:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.mkv
4916 Display CIE color diagram with pixels overlaid onto it.
4918 The filter accepts the following options:
4933 @item uhdtv, rec2020
4946 Set what gamuts to draw.
4948 See @code{system} option for available values.
4951 Set ciescope size, by default set to 512.
4954 Set intensity used to map input pixel values to CIE diagram.
4957 Set contrast used to draw tongue colors that are out of active color system gamut.
4960 Correct gamma displayed on scope, by default enabled.
4963 Show white point on CIE diagram, by default disabled.
4966 Set input gamma. Used only with XYZ input color space.
4971 Visualize information exported by some codecs.
4973 Some codecs can export information through frames using side-data or other
4974 means. For example, some MPEG based codecs export motion vectors through the
4975 @var{export_mvs} flag in the codec @option{flags2} option.
4977 The filter accepts the following option:
4981 Set motion vectors to visualize.
4983 Available flags for @var{mv} are:
4987 forward predicted MVs of P-frames
4989 forward predicted MVs of B-frames
4991 backward predicted MVs of B-frames
4995 Display quantization parameters using the chroma planes.
4998 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
5000 Available flags for @var{mv_type} are:
5004 forward predicted MVs
5006 backward predicted MVs
5009 @item frame_type, ft
5010 Set frame type to visualize motion vectors of.
5012 Available flags for @var{frame_type} are:
5016 intra-coded frames (I-frames)
5018 predicted frames (P-frames)
5020 bi-directionally predicted frames (B-frames)
5024 @subsection Examples
5028 Visualize forward predicted MVs of all frames using @command{ffplay}:
5030 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
5034 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5036 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5040 @section colorbalance
5041 Modify intensity of primary colors (red, green and blue) of input frames.
5043 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5044 regions for the red-cyan, green-magenta or blue-yellow balance.
5046 A positive adjustment value shifts the balance towards the primary color, a negative
5047 value towards the complementary color.
5049 The filter accepts the following options:
5055 Adjust red, green and blue shadows (darkest pixels).
5060 Adjust red, green and blue midtones (medium pixels).
5065 Adjust red, green and blue highlights (brightest pixels).
5067 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5070 @subsection Examples
5074 Add red color cast to shadows:
5081 RGB colorspace color keying.
5083 The filter accepts the following options:
5087 The color which will be replaced with transparency.
5090 Similarity percentage with the key color.
5092 0.01 matches only the exact key color, while 1.0 matches everything.
5097 0.0 makes pixels either fully transparent, or not transparent at all.
5099 Higher values result in semi-transparent pixels, with a higher transparency
5100 the more similar the pixels color is to the key color.
5103 @subsection Examples
5107 Make every green pixel in the input image transparent:
5109 ffmpeg -i input.png -vf colorkey=green out.png
5113 Overlay a greenscreen-video on top of a static background image.
5115 ffmpeg -i background.png -i video.mp4 -filter_complex "[1:v]colorkey=0x3BBD1E:0.3:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.flv
5119 @section colorlevels
5121 Adjust video input frames using levels.
5123 The filter accepts the following options:
5130 Adjust red, green, blue and alpha input black point.
5131 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5137 Adjust red, green, blue and alpha input white point.
5138 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
5140 Input levels are used to lighten highlights (bright tones), darken shadows
5141 (dark tones), change the balance of bright and dark tones.
5147 Adjust red, green, blue and alpha output black point.
5148 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5154 Adjust red, green, blue and alpha output white point.
5155 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5157 Output levels allows manual selection of a constrained output level range.
5160 @subsection Examples
5164 Make video output darker:
5166 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5172 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5176 Make video output lighter:
5178 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5182 Increase brightness:
5184 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5188 @section colorchannelmixer
5190 Adjust video input frames by re-mixing color channels.
5192 This filter modifies a color channel by adding the values associated to
5193 the other channels of the same pixels. For example if the value to
5194 modify is red, the output value will be:
5196 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5199 The filter accepts the following options:
5206 Adjust contribution of input red, green, blue and alpha channels for output red channel.
5207 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5213 Adjust contribution of input red, green, blue and alpha channels for output green channel.
5214 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5220 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5221 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5227 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5228 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5230 Allowed ranges for options are @code{[-2.0, 2.0]}.
5233 @subsection Examples
5237 Convert source to grayscale:
5239 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5242 Simulate sepia tones:
5244 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5248 @section colormatrix
5250 Convert color matrix.
5252 The filter accepts the following options:
5257 Specify the source and destination color matrix. Both values must be
5260 The accepted values are:
5288 For example to convert from BT.601 to SMPTE-240M, use the command:
5290 colormatrix=bt601:smpte240m
5295 Convert colorspace, transfer characteristics or color primaries.
5296 Input video needs to have an even size.
5298 The filter accepts the following options:
5303 Specify all color properties at once.
5305 The accepted values are:
5335 Specify output colorspace.
5337 The accepted values are:
5346 BT.470BG or BT.601-6 625
5349 SMPTE-170M or BT.601-6 525
5358 BT.2020 with non-constant luminance
5364 Specify output transfer characteristics.
5366 The accepted values are:
5378 Constant gamma of 2.2
5381 Constant gamma of 2.8
5384 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5402 BT.2020 for 10-bits content
5405 BT.2020 for 12-bits content
5411 Specify output color primaries.
5413 The accepted values are:
5422 BT.470BG or BT.601-6 625
5425 SMPTE-170M or BT.601-6 525
5446 Specify output color range.
5448 The accepted values are:
5451 TV (restricted) range
5454 MPEG (restricted) range
5465 Specify output color format.
5467 The accepted values are:
5470 YUV 4:2:0 planar 8-bits
5473 YUV 4:2:0 planar 10-bits
5476 YUV 4:2:0 planar 12-bits
5479 YUV 4:2:2 planar 8-bits
5482 YUV 4:2:2 planar 10-bits
5485 YUV 4:2:2 planar 12-bits
5488 YUV 4:4:4 planar 8-bits
5491 YUV 4:4:4 planar 10-bits
5494 YUV 4:4:4 planar 12-bits
5499 Do a fast conversion, which skips gamma/primary correction. This will take
5500 significantly less CPU, but will be mathematically incorrect. To get output
5501 compatible with that produced by the colormatrix filter, use fast=1.
5504 Specify dithering mode.
5506 The accepted values are:
5512 Floyd-Steinberg dithering
5516 Whitepoint adaptation mode.
5518 The accepted values are:
5521 Bradford whitepoint adaptation
5524 von Kries whitepoint adaptation
5527 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5531 Override all input properties at once. Same accepted values as @ref{all}.
5534 Override input colorspace. Same accepted values as @ref{space}.
5537 Override input color primaries. Same accepted values as @ref{primaries}.
5540 Override input transfer characteristics. Same accepted values as @ref{trc}.
5543 Override input color range. Same accepted values as @ref{range}.
5547 The filter converts the transfer characteristics, color space and color
5548 primaries to the specified user values. The output value, if not specified,
5549 is set to a default value based on the "all" property. If that property is
5550 also not specified, the filter will log an error. The output color range and
5551 format default to the same value as the input color range and format. The
5552 input transfer characteristics, color space, color primaries and color range
5553 should be set on the input data. If any of these are missing, the filter will
5554 log an error and no conversion will take place.
5556 For example to convert the input to SMPTE-240M, use the command:
5558 colorspace=smpte240m
5561 @section convolution
5563 Apply convolution 3x3 or 5x5 filter.
5565 The filter accepts the following options:
5572 Set matrix for each plane.
5573 Matrix is sequence of 9 or 25 signed integers.
5579 Set multiplier for calculated value for each plane.
5585 Set bias for each plane. This value is added to the result of the multiplication.
5586 Useful for making the overall image brighter or darker. Default is 0.0.
5589 @subsection Examples
5595 convolution="0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0"
5601 convolution="1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1/9:1/9:1/9:1/9"
5607 convolution="0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:5:1:1:1:0:128:128:128"
5613 convolution="0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:5:5:5:1:0:128:128:128"
5619 convolution="-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2"
5625 Copy the input source unchanged to the output. This is mainly useful for
5630 Video filtering on GPU using Apple's CoreImage API on OSX.
5632 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5633 processed by video hardware. However, software-based OpenGL implementations
5634 exist which means there is no guarantee for hardware processing. It depends on
5637 There are many filters and image generators provided by Apple that come with a
5638 large variety of options. The filter has to be referenced by its name along
5641 The coreimage filter accepts the following options:
5644 List all available filters and generators along with all their respective
5645 options as well as possible minimum and maximum values along with the default
5652 Specify all filters by their respective name and options.
5653 Use @var{list_filters} to determine all valid filter names and options.
5654 Numerical options are specified by a float value and are automatically clamped
5655 to their respective value range. Vector and color options have to be specified
5656 by a list of space separated float values. Character escaping has to be done.
5657 A special option name @code{default} is available to use default options for a
5660 It is required to specify either @code{default} or at least one of the filter options.
5661 All omitted options are used with their default values.
5662 The syntax of the filter string is as follows:
5664 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5668 Specify a rectangle where the output of the filter chain is copied into the
5669 input image. It is given by a list of space separated float values:
5671 output_rect=x\ y\ width\ height
5673 If not given, the output rectangle equals the dimensions of the input image.
5674 The output rectangle is automatically cropped at the borders of the input
5675 image. Negative values are valid for each component.
5677 output_rect=25\ 25\ 100\ 100
5681 Several filters can be chained for successive processing without GPU-HOST
5682 transfers allowing for fast processing of complex filter chains.
5683 Currently, only filters with zero (generators) or exactly one (filters) input
5684 image and one output image are supported. Also, transition filters are not yet
5687 Some filters generate output images with additional padding depending on the
5688 respective filter kernel. The padding is automatically removed to ensure the
5689 filter output has the same size as the input image.
5691 For image generators, the size of the output image is determined by the
5692 previous output image of the filter chain or the input image of the whole
5693 filterchain, respectively. The generators do not use the pixel information of
5694 this image to generate their output. However, the generated output is
5695 blended onto this image, resulting in partial or complete coverage of the
5698 The @ref{coreimagesrc} video source can be used for generating input images
5699 which are directly fed into the filter chain. By using it, providing input
5700 images by another video source or an input video is not required.
5702 @subsection Examples
5707 List all filters available:
5709 coreimage=list_filters=true
5713 Use the CIBoxBlur filter with default options to blur an image:
5715 coreimage=filter=CIBoxBlur@@default
5719 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5720 its center at 100x100 and a radius of 50 pixels:
5722 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5726 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5727 given as complete and escaped command-line for Apple's standard bash shell:
5729 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5735 Crop the input video to given dimensions.
5737 It accepts the following parameters:
5741 The width of the output video. It defaults to @code{iw}.
5742 This expression is evaluated only once during the filter
5743 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5746 The height of the output video. It defaults to @code{ih}.
5747 This expression is evaluated only once during the filter
5748 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5751 The horizontal position, in the input video, of the left edge of the output
5752 video. It defaults to @code{(in_w-out_w)/2}.
5753 This expression is evaluated per-frame.
5756 The vertical position, in the input video, of the top edge of the output video.
5757 It defaults to @code{(in_h-out_h)/2}.
5758 This expression is evaluated per-frame.
5761 If set to 1 will force the output display aspect ratio
5762 to be the same of the input, by changing the output sample aspect
5763 ratio. It defaults to 0.
5766 Enable exact cropping. If enabled, subsampled videos will be cropped at exact
5767 width/height/x/y as specified and will not be rounded to nearest smaller value.
5771 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5772 expressions containing the following constants:
5777 The computed values for @var{x} and @var{y}. They are evaluated for
5782 The input width and height.
5786 These are the same as @var{in_w} and @var{in_h}.
5790 The output (cropped) width and height.
5794 These are the same as @var{out_w} and @var{out_h}.
5797 same as @var{iw} / @var{ih}
5800 input sample aspect ratio
5803 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5807 horizontal and vertical chroma subsample values. For example for the
5808 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5811 The number of the input frame, starting from 0.
5814 the position in the file of the input frame, NAN if unknown
5817 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5821 The expression for @var{out_w} may depend on the value of @var{out_h},
5822 and the expression for @var{out_h} may depend on @var{out_w}, but they
5823 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5824 evaluated after @var{out_w} and @var{out_h}.
5826 The @var{x} and @var{y} parameters specify the expressions for the
5827 position of the top-left corner of the output (non-cropped) area. They
5828 are evaluated for each frame. If the evaluated value is not valid, it
5829 is approximated to the nearest valid value.
5831 The expression for @var{x} may depend on @var{y}, and the expression
5832 for @var{y} may depend on @var{x}.
5834 @subsection Examples
5838 Crop area with size 100x100 at position (12,34).
5843 Using named options, the example above becomes:
5845 crop=w=100:h=100:x=12:y=34
5849 Crop the central input area with size 100x100:
5855 Crop the central input area with size 2/3 of the input video:
5857 crop=2/3*in_w:2/3*in_h
5861 Crop the input video central square:
5868 Delimit the rectangle with the top-left corner placed at position
5869 100:100 and the right-bottom corner corresponding to the right-bottom
5870 corner of the input image.
5872 crop=in_w-100:in_h-100:100:100
5876 Crop 10 pixels from the left and right borders, and 20 pixels from
5877 the top and bottom borders
5879 crop=in_w-2*10:in_h-2*20
5883 Keep only the bottom right quarter of the input image:
5885 crop=in_w/2:in_h/2:in_w/2:in_h/2
5889 Crop height for getting Greek harmony:
5891 crop=in_w:1/PHI*in_w
5895 Apply trembling effect:
5897 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)
5901 Apply erratic camera effect depending on timestamp:
5903 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"
5907 Set x depending on the value of y:
5909 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5913 @subsection Commands
5915 This filter supports the following commands:
5921 Set width/height of the output video and the horizontal/vertical position
5923 The command accepts the same syntax of the corresponding option.
5925 If the specified expression is not valid, it is kept at its current
5931 Auto-detect the crop size.
5933 It calculates the necessary cropping parameters and prints the
5934 recommended parameters via the logging system. The detected dimensions
5935 correspond to the non-black area of the input video.
5937 It accepts the following parameters:
5942 Set higher black value threshold, which can be optionally specified
5943 from nothing (0) to everything (255 for 8-bit based formats). An intensity
5944 value greater to the set value is considered non-black. It defaults to 24.
5945 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5946 on the bitdepth of the pixel format.
5949 The value which the width/height should be divisible by. It defaults to
5950 16. The offset is automatically adjusted to center the video. Use 2 to
5951 get only even dimensions (needed for 4:2:2 video). 16 is best when
5952 encoding to most video codecs.
5954 @item reset_count, reset
5955 Set the counter that determines after how many frames cropdetect will
5956 reset the previously detected largest video area and start over to
5957 detect the current optimal crop area. Default value is 0.
5959 This can be useful when channel logos distort the video area. 0
5960 indicates 'never reset', and returns the largest area encountered during
5967 Apply color adjustments using curves.
5969 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5970 component (red, green and blue) has its values defined by @var{N} key points
5971 tied from each other using a smooth curve. The x-axis represents the pixel
5972 values from the input frame, and the y-axis the new pixel values to be set for
5975 By default, a component curve is defined by the two points @var{(0;0)} and
5976 @var{(1;1)}. This creates a straight line where each original pixel value is
5977 "adjusted" to its own value, which means no change to the image.
5979 The filter allows you to redefine these two points and add some more. A new
5980 curve (using a natural cubic spline interpolation) will be define to pass
5981 smoothly through all these new coordinates. The new defined points needs to be
5982 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5983 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5984 the vector spaces, the values will be clipped accordingly.
5986 The filter accepts the following options:
5990 Select one of the available color presets. This option can be used in addition
5991 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5992 options takes priority on the preset values.
5993 Available presets are:
5996 @item color_negative
5999 @item increase_contrast
6001 @item linear_contrast
6002 @item medium_contrast
6004 @item strong_contrast
6007 Default is @code{none}.
6009 Set the master key points. These points will define a second pass mapping. It
6010 is sometimes called a "luminance" or "value" mapping. It can be used with
6011 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
6012 post-processing LUT.
6014 Set the key points for the red component.
6016 Set the key points for the green component.
6018 Set the key points for the blue component.
6020 Set the key points for all components (not including master).
6021 Can be used in addition to the other key points component
6022 options. In this case, the unset component(s) will fallback on this
6023 @option{all} setting.
6025 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
6027 Save Gnuplot script of the curves in specified file.
6030 To avoid some filtergraph syntax conflicts, each key points list need to be
6031 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
6033 @subsection Examples
6037 Increase slightly the middle level of blue:
6039 curves=blue='0/0 0.5/0.58 1/1'
6045 curves=r='0/0.11 .42/.51 1/0.95':g='0/0 0.50/0.48 1/1':b='0/0.22 .49/.44 1/0.8'
6047 Here we obtain the following coordinates for each components:
6050 @code{(0;0.11) (0.42;0.51) (1;0.95)}
6052 @code{(0;0) (0.50;0.48) (1;1)}
6054 @code{(0;0.22) (0.49;0.44) (1;0.80)}
6058 The previous example can also be achieved with the associated built-in preset:
6060 curves=preset=vintage
6070 Use a Photoshop preset and redefine the points of the green component:
6072 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6076 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6077 and @command{gnuplot}:
6079 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
6080 gnuplot -p /tmp/curves.plt
6086 Video data analysis filter.
6088 This filter shows hexadecimal pixel values of part of video.
6090 The filter accepts the following options:
6094 Set output video size.
6097 Set x offset from where to pick pixels.
6100 Set y offset from where to pick pixels.
6103 Set scope mode, can be one of the following:
6106 Draw hexadecimal pixel values with white color on black background.
6109 Draw hexadecimal pixel values with input video pixel color on black
6113 Draw hexadecimal pixel values on color background picked from input video,
6114 the text color is picked in such way so its always visible.
6118 Draw rows and columns numbers on left and top of video.
6121 Set background opacity.
6126 Denoise frames using 2D DCT (frequency domain filtering).
6128 This filter is not designed for real time.
6130 The filter accepts the following options:
6134 Set the noise sigma constant.
6136 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6137 coefficient (absolute value) below this threshold with be dropped.
6139 If you need a more advanced filtering, see @option{expr}.
6141 Default is @code{0}.
6144 Set number overlapping pixels for each block. Since the filter can be slow, you
6145 may want to reduce this value, at the cost of a less effective filter and the
6146 risk of various artefacts.
6148 If the overlapping value doesn't permit processing the whole input width or
6149 height, a warning will be displayed and according borders won't be denoised.
6151 Default value is @var{blocksize}-1, which is the best possible setting.
6154 Set the coefficient factor expression.
6156 For each coefficient of a DCT block, this expression will be evaluated as a
6157 multiplier value for the coefficient.
6159 If this is option is set, the @option{sigma} option will be ignored.
6161 The absolute value of the coefficient can be accessed through the @var{c}
6165 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
6166 @var{blocksize}, which is the width and height of the processed blocks.
6168 The default value is @var{3} (8x8) and can be raised to @var{4} for a
6169 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
6170 on the speed processing. Also, a larger block size does not necessarily means a
6174 @subsection Examples
6176 Apply a denoise with a @option{sigma} of @code{4.5}:
6181 The same operation can be achieved using the expression system:
6183 dctdnoiz=e='gte(c, 4.5*3)'
6186 Violent denoise using a block size of @code{16x16}:
6193 Remove banding artifacts from input video.
6194 It works by replacing banded pixels with average value of referenced pixels.
6196 The filter accepts the following options:
6203 Set banding detection threshold for each plane. Default is 0.02.
6204 Valid range is 0.00003 to 0.5.
6205 If difference between current pixel and reference pixel is less than threshold,
6206 it will be considered as banded.
6209 Banding detection range in pixels. Default is 16. If positive, random number
6210 in range 0 to set value will be used. If negative, exact absolute value
6212 The range defines square of four pixels around current pixel.
6215 Set direction in radians from which four pixel will be compared. If positive,
6216 random direction from 0 to set direction will be picked. If negative, exact of
6217 absolute value will be picked. For example direction 0, -PI or -2*PI radians
6218 will pick only pixels on same row and -PI/2 will pick only pixels on same
6222 If enabled, current pixel is compared with average value of all four
6223 surrounding pixels. The default is enabled. If disabled current pixel is
6224 compared with all four surrounding pixels. The pixel is considered banded
6225 if only all four differences with surrounding pixels are less than threshold.
6228 If enabled, current pixel is changed if and only if all pixel components are banded,
6229 e.g. banding detection threshold is triggered for all color components.
6230 The default is disabled.
6236 Drop duplicated frames at regular intervals.
6238 The filter accepts the following options:
6242 Set the number of frames from which one will be dropped. Setting this to
6243 @var{N} means one frame in every batch of @var{N} frames will be dropped.
6244 Default is @code{5}.
6247 Set the threshold for duplicate detection. If the difference metric for a frame
6248 is less than or equal to this value, then it is declared as duplicate. Default
6252 Set scene change threshold. Default is @code{15}.
6256 Set the size of the x and y-axis blocks used during metric calculations.
6257 Larger blocks give better noise suppression, but also give worse detection of
6258 small movements. Must be a power of two. Default is @code{32}.
6261 Mark main input as a pre-processed input and activate clean source input
6262 stream. This allows the input to be pre-processed with various filters to help
6263 the metrics calculation while keeping the frame selection lossless. When set to
6264 @code{1}, the first stream is for the pre-processed input, and the second
6265 stream is the clean source from where the kept frames are chosen. Default is
6269 Set whether or not chroma is considered in the metric calculations. Default is
6275 Apply deflate effect to the video.
6277 This filter replaces the pixel by the local(3x3) average by taking into account
6278 only values lower than the pixel.
6280 It accepts the following options:
6287 Limit the maximum change for each plane, default is 65535.
6288 If 0, plane will remain unchanged.
6293 Remove temporal frame luminance variations.
6295 It accepts the following options:
6299 Set moving-average filter size in frames. Default is 5. Allowed range is 2 - 129.
6302 Set averaging mode to smooth temporal luminance variations.
6304 Available values are:
6331 Remove judder produced by partially interlaced telecined content.
6333 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6334 source was partially telecined content then the output of @code{pullup,dejudder}
6335 will have a variable frame rate. May change the recorded frame rate of the
6336 container. Aside from that change, this filter will not affect constant frame
6339 The option available in this filter is:
6343 Specify the length of the window over which the judder repeats.
6345 Accepts any integer greater than 1. Useful values are:
6349 If the original was telecined from 24 to 30 fps (Film to NTSC).
6352 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6355 If a mixture of the two.
6358 The default is @samp{4}.
6363 Suppress a TV station logo by a simple interpolation of the surrounding
6364 pixels. Just set a rectangle covering the logo and watch it disappear
6365 (and sometimes something even uglier appear - your mileage may vary).
6367 It accepts the following parameters:
6372 Specify the top left corner coordinates of the logo. They must be
6377 Specify the width and height of the logo to clear. They must be
6381 Specify the thickness of the fuzzy edge of the rectangle (added to
6382 @var{w} and @var{h}). The default value is 1. This option is
6383 deprecated, setting higher values should no longer be necessary and
6387 When set to 1, a green rectangle is drawn on the screen to simplify
6388 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6389 The default value is 0.
6391 The rectangle is drawn on the outermost pixels which will be (partly)
6392 replaced with interpolated values. The values of the next pixels
6393 immediately outside this rectangle in each direction will be used to
6394 compute the interpolated pixel values inside the rectangle.
6398 @subsection Examples
6402 Set a rectangle covering the area with top left corner coordinates 0,0
6403 and size 100x77, and a band of size 10:
6405 delogo=x=0:y=0:w=100:h=77:band=10
6412 Attempt to fix small changes in horizontal and/or vertical shift. This
6413 filter helps remove camera shake from hand-holding a camera, bumping a
6414 tripod, moving on a vehicle, etc.
6416 The filter accepts the following options:
6424 Specify a rectangular area where to limit the search for motion
6426 If desired the search for motion vectors can be limited to a
6427 rectangular area of the frame defined by its top left corner, width
6428 and height. These parameters have the same meaning as the drawbox
6429 filter which can be used to visualise the position of the bounding
6432 This is useful when simultaneous movement of subjects within the frame
6433 might be confused for camera motion by the motion vector search.
6435 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6436 then the full frame is used. This allows later options to be set
6437 without specifying the bounding box for the motion vector search.
6439 Default - search the whole frame.
6443 Specify the maximum extent of movement in x and y directions in the
6444 range 0-64 pixels. Default 16.
6447 Specify how to generate pixels to fill blanks at the edge of the
6448 frame. Available values are:
6451 Fill zeroes at blank locations
6453 Original image at blank locations
6455 Extruded edge value at blank locations
6457 Mirrored edge at blank locations
6459 Default value is @samp{mirror}.
6462 Specify the blocksize to use for motion search. Range 4-128 pixels,
6466 Specify the contrast threshold for blocks. Only blocks with more than
6467 the specified contrast (difference between darkest and lightest
6468 pixels) will be considered. Range 1-255, default 125.
6471 Specify the search strategy. Available values are:
6474 Set exhaustive search
6476 Set less exhaustive search.
6478 Default value is @samp{exhaustive}.
6481 If set then a detailed log of the motion search is written to the
6485 If set to 1, specify using OpenCL capabilities, only available if
6486 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6492 Apply an exact inverse of the telecine operation. It requires a predefined
6493 pattern specified using the pattern option which must be the same as that passed
6494 to the telecine filter.
6496 This filter accepts the following options:
6505 The default value is @code{top}.
6509 A string of numbers representing the pulldown pattern you wish to apply.
6510 The default value is @code{23}.
6513 A number representing position of the first frame with respect to the telecine
6514 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6519 Apply dilation effect to the video.
6521 This filter replaces the pixel by the local(3x3) maximum.
6523 It accepts the following options:
6530 Limit the maximum change for each plane, default is 65535.
6531 If 0, plane will remain unchanged.
6534 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6537 Flags to local 3x3 coordinates maps like this:
6546 Displace pixels as indicated by second and third input stream.
6548 It takes three input streams and outputs one stream, the first input is the
6549 source, and second and third input are displacement maps.
6551 The second input specifies how much to displace pixels along the
6552 x-axis, while the third input specifies how much to displace pixels
6554 If one of displacement map streams terminates, last frame from that
6555 displacement map will be used.
6557 Note that once generated, displacements maps can be reused over and over again.
6559 A description of the accepted options follows.
6563 Set displace behavior for pixels that are out of range.
6565 Available values are:
6568 Missing pixels are replaced by black pixels.
6571 Adjacent pixels will spread out to replace missing pixels.
6574 Out of range pixels are wrapped so they point to pixels of other side.
6576 Default is @samp{smear}.
6580 @subsection Examples
6584 Add ripple effect to rgb input of video size hd720:
6586 ffmpeg -i INPUT -f lavfi -i nullsrc=s=hd720,lutrgb=128:128:128 -f lavfi -i nullsrc=s=hd720,geq='r=128+30*sin(2*PI*X/400+T):g=128+30*sin(2*PI*X/400+T):b=128+30*sin(2*PI*X/400+T)' -lavfi '[0][1][2]displace' OUTPUT
6590 Add wave effect to rgb input of video size hd720:
6592 ffmpeg -i INPUT -f lavfi -i nullsrc=hd720,geq='r=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):g=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):b=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T))' -lavfi '[1]split[x][y],[0][x][y]displace' OUTPUT
6598 Draw a colored box on the input image.
6600 It accepts the following parameters:
6605 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6609 The expressions which specify the width and height of the box; if 0 they are interpreted as
6610 the input width and height. It defaults to 0.
6613 Specify the color of the box to write. For the general syntax of this option,
6614 check the "Color" section in the ffmpeg-utils manual. If the special
6615 value @code{invert} is used, the box edge color is the same as the
6616 video with inverted luma.
6619 The expression which sets the thickness of the box edge. Default value is @code{3}.
6621 See below for the list of accepted constants.
6624 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6625 following constants:
6629 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6633 horizontal and vertical chroma subsample values. For example for the
6634 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6638 The input width and height.
6641 The input sample aspect ratio.
6645 The x and y offset coordinates where the box is drawn.
6649 The width and height of the drawn box.
6652 The thickness of the drawn box.
6654 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6655 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6659 @subsection Examples
6663 Draw a black box around the edge of the input image:
6669 Draw a box with color red and an opacity of 50%:
6671 drawbox=10:20:200:60:red@@0.5
6674 The previous example can be specified as:
6676 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6680 Fill the box with pink color:
6682 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6686 Draw a 2-pixel red 2.40:1 mask:
6688 drawbox=x=-t:y=0.5*(ih-iw/2.4)-t:w=iw+t*2:h=iw/2.4+t*2:t=2:c=red
6694 Draw a grid on the input image.
6696 It accepts the following parameters:
6701 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6705 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6706 input width and height, respectively, minus @code{thickness}, so image gets
6707 framed. Default to 0.
6710 Specify the color of the grid. For the general syntax of this option,
6711 check the "Color" section in the ffmpeg-utils manual. If the special
6712 value @code{invert} is used, the grid color is the same as the
6713 video with inverted luma.
6716 The expression which sets the thickness of the grid line. Default value is @code{1}.
6718 See below for the list of accepted constants.
6721 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6722 following constants:
6726 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6730 horizontal and vertical chroma subsample values. For example for the
6731 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6735 The input grid cell width and height.
6738 The input sample aspect ratio.
6742 The x and y coordinates of some point of grid intersection (meant to configure offset).
6746 The width and height of the drawn cell.
6749 The thickness of the drawn cell.
6751 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6752 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6756 @subsection Examples
6760 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6762 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6766 Draw a white 3x3 grid with an opacity of 50%:
6768 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6775 Draw a text string or text from a specified file on top of a video, using the
6776 libfreetype library.
6778 To enable compilation of this filter, you need to configure FFmpeg with
6779 @code{--enable-libfreetype}.
6780 To enable default font fallback and the @var{font} option you need to
6781 configure FFmpeg with @code{--enable-libfontconfig}.
6782 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6783 @code{--enable-libfribidi}.
6787 It accepts the following parameters:
6792 Used to draw a box around text using the background color.
6793 The value must be either 1 (enable) or 0 (disable).
6794 The default value of @var{box} is 0.
6797 Set the width of the border to be drawn around the box using @var{boxcolor}.
6798 The default value of @var{boxborderw} is 0.
6801 The color to be used for drawing box around text. For the syntax of this
6802 option, check the "Color" section in the ffmpeg-utils manual.
6804 The default value of @var{boxcolor} is "white".
6807 Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6808 The default value of @var{line_spacing} is 0.
6811 Set the width of the border to be drawn around the text using @var{bordercolor}.
6812 The default value of @var{borderw} is 0.
6815 Set the color to be used for drawing border around text. For the syntax of this
6816 option, check the "Color" section in the ffmpeg-utils manual.
6818 The default value of @var{bordercolor} is "black".
6821 Select how the @var{text} is expanded. Can be either @code{none},
6822 @code{strftime} (deprecated) or
6823 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6827 Set a start time for the count. Value is in microseconds. Only applied
6828 in the deprecated strftime expansion mode. To emulate in normal expansion
6829 mode use the @code{pts} function, supplying the start time (in seconds)
6830 as the second argument.
6833 If true, check and fix text coords to avoid clipping.
6836 The color to be used for drawing fonts. For the syntax of this option, check
6837 the "Color" section in the ffmpeg-utils manual.
6839 The default value of @var{fontcolor} is "black".
6841 @item fontcolor_expr
6842 String which is expanded the same way as @var{text} to obtain dynamic
6843 @var{fontcolor} value. By default this option has empty value and is not
6844 processed. When this option is set, it overrides @var{fontcolor} option.
6847 The font family to be used for drawing text. By default Sans.
6850 The font file to be used for drawing text. The path must be included.
6851 This parameter is mandatory if the fontconfig support is disabled.
6854 Draw the text applying alpha blending. The value can
6855 be a number between 0.0 and 1.0.
6856 The expression accepts the same variables @var{x, y} as well.
6857 The default value is 1.
6858 Please see @var{fontcolor_expr}.
6861 The font size to be used for drawing text.
6862 The default value of @var{fontsize} is 16.
6865 If set to 1, attempt to shape the text (for example, reverse the order of
6866 right-to-left text and join Arabic characters) before drawing it.
6867 Otherwise, just draw the text exactly as given.
6868 By default 1 (if supported).
6871 The flags to be used for loading the fonts.
6873 The flags map the corresponding flags supported by libfreetype, and are
6874 a combination of the following values:
6881 @item vertical_layout
6882 @item force_autohint
6885 @item ignore_global_advance_width
6887 @item ignore_transform
6893 Default value is "default".
6895 For more information consult the documentation for the FT_LOAD_*
6899 The color to be used for drawing a shadow behind the drawn text. For the
6900 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6902 The default value of @var{shadowcolor} is "black".
6906 The x and y offsets for the text shadow position with respect to the
6907 position of the text. They can be either positive or negative
6908 values. The default value for both is "0".
6911 The starting frame number for the n/frame_num variable. The default value
6915 The size in number of spaces to use for rendering the tab.
6919 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6920 format. It can be used with or without text parameter. @var{timecode_rate}
6921 option must be specified.
6923 @item timecode_rate, rate, r
6924 Set the timecode frame rate (timecode only).
6927 If set to 1, the output of the timecode option will wrap around at 24 hours.
6928 Default is 0 (disabled).
6931 The text string to be drawn. The text must be a sequence of UTF-8
6933 This parameter is mandatory if no file is specified with the parameter
6937 A text file containing text to be drawn. The text must be a sequence
6938 of UTF-8 encoded characters.
6940 This parameter is mandatory if no text string is specified with the
6941 parameter @var{text}.
6943 If both @var{text} and @var{textfile} are specified, an error is thrown.
6946 If set to 1, the @var{textfile} will be reloaded before each frame.
6947 Be sure to update it atomically, or it may be read partially, or even fail.
6951 The expressions which specify the offsets where text will be drawn
6952 within the video frame. They are relative to the top/left border of the
6955 The default value of @var{x} and @var{y} is "0".
6957 See below for the list of accepted constants and functions.
6960 The parameters for @var{x} and @var{y} are expressions containing the
6961 following constants and functions:
6965 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6969 horizontal and vertical chroma subsample values. For example for the
6970 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6973 the height of each text line
6981 @item max_glyph_a, ascent
6982 the maximum distance from the baseline to the highest/upper grid
6983 coordinate used to place a glyph outline point, for all the rendered
6985 It is a positive value, due to the grid's orientation with the Y axis
6988 @item max_glyph_d, descent
6989 the maximum distance from the baseline to the lowest grid coordinate
6990 used to place a glyph outline point, for all the rendered glyphs.
6991 This is a negative value, due to the grid's orientation, with the Y axis
6995 maximum glyph height, that is the maximum height for all the glyphs
6996 contained in the rendered text, it is equivalent to @var{ascent} -
7000 maximum glyph width, that is the maximum width for all the glyphs
7001 contained in the rendered text
7004 the number of input frame, starting from 0
7006 @item rand(min, max)
7007 return a random number included between @var{min} and @var{max}
7010 The input sample aspect ratio.
7013 timestamp expressed in seconds, NAN if the input timestamp is unknown
7016 the height of the rendered text
7019 the width of the rendered text
7023 the x and y offset coordinates where the text is drawn.
7025 These parameters allow the @var{x} and @var{y} expressions to refer
7026 each other, so you can for example specify @code{y=x/dar}.
7029 @anchor{drawtext_expansion}
7030 @subsection Text expansion
7032 If @option{expansion} is set to @code{strftime},
7033 the filter recognizes strftime() sequences in the provided text and
7034 expands them accordingly. Check the documentation of strftime(). This
7035 feature is deprecated.
7037 If @option{expansion} is set to @code{none}, the text is printed verbatim.
7039 If @option{expansion} is set to @code{normal} (which is the default),
7040 the following expansion mechanism is used.
7042 The backslash character @samp{\}, followed by any character, always expands to
7043 the second character.
7045 Sequences of the form @code{%@{...@}} are expanded. The text between the
7046 braces is a function name, possibly followed by arguments separated by ':'.
7047 If the arguments contain special characters or delimiters (':' or '@}'),
7048 they should be escaped.
7050 Note that they probably must also be escaped as the value for the
7051 @option{text} option in the filter argument string and as the filter
7052 argument in the filtergraph description, and possibly also for the shell,
7053 that makes up to four levels of escaping; using a text file avoids these
7056 The following functions are available:
7061 The expression evaluation result.
7063 It must take one argument specifying the expression to be evaluated,
7064 which accepts the same constants and functions as the @var{x} and
7065 @var{y} values. Note that not all constants should be used, for
7066 example the text size is not known when evaluating the expression, so
7067 the constants @var{text_w} and @var{text_h} will have an undefined
7070 @item expr_int_format, eif
7071 Evaluate the expression's value and output as formatted integer.
7073 The first argument is the expression to be evaluated, just as for the @var{expr} function.
7074 The second argument specifies the output format. Allowed values are @samp{x},
7075 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7076 @code{printf} function.
7077 The third parameter is optional and sets the number of positions taken by the output.
7078 It can be used to add padding with zeros from the left.
7081 The time at which the filter is running, expressed in UTC.
7082 It can accept an argument: a strftime() format string.
7085 The time at which the filter is running, expressed in the local time zone.
7086 It can accept an argument: a strftime() format string.
7089 Frame metadata. Takes one or two arguments.
7091 The first argument is mandatory and specifies the metadata key.
7093 The second argument is optional and specifies a default value, used when the
7094 metadata key is not found or empty.
7097 The frame number, starting from 0.
7100 A 1 character description of the current picture type.
7103 The timestamp of the current frame.
7104 It can take up to three arguments.
7106 The first argument is the format of the timestamp; it defaults to @code{flt}
7107 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7108 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7109 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7110 @code{localtime} stands for the timestamp of the frame formatted as
7111 local time zone time.
7113 The second argument is an offset added to the timestamp.
7115 If the format is set to @code{localtime} or @code{gmtime},
7116 a third argument may be supplied: a strftime() format string.
7117 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7120 @subsection Examples
7124 Draw "Test Text" with font FreeSerif, using the default values for the
7125 optional parameters.
7128 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7132 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7133 and y=50 (counting from the top-left corner of the screen), text is
7134 yellow with a red box around it. Both the text and the box have an
7138 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7139 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7142 Note that the double quotes are not necessary if spaces are not used
7143 within the parameter list.
7146 Show the text at the center of the video frame:
7148 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7152 Show the text at a random position, switching to a new position every 30 seconds:
7154 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=if(eq(mod(t\,30)\,0)\,rand(0\,(w-text_w))\,x):y=if(eq(mod(t\,30)\,0)\,rand(0\,(h-text_h))\,y)"
7158 Show a text line sliding from right to left in the last row of the video
7159 frame. The file @file{LONG_LINE} is assumed to contain a single line
7162 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7166 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7168 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7172 Draw a single green letter "g", at the center of the input video.
7173 The glyph baseline is placed at half screen height.
7175 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7179 Show text for 1 second every 3 seconds:
7181 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7185 Use fontconfig to set the font. Note that the colons need to be escaped.
7187 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7191 Print the date of a real-time encoding (see strftime(3)):
7193 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7197 Show text fading in and out (appearing/disappearing):
7200 DS=1.0 # display start
7201 DE=10.0 # display end
7202 FID=1.5 # fade in duration
7203 FOD=5 # fade out duration
7204 ffplay -f lavfi "color,drawtext=text=TEST:fontsize=50:fontfile=FreeSerif.ttf:fontcolor_expr=ff0000%@{eif\\\\: clip(255*(1*between(t\\, $DS + $FID\\, $DE - $FOD) + ((t - $DS)/$FID)*between(t\\, $DS\\, $DS + $FID) + (-(t - $DE)/$FOD)*between(t\\, $DE - $FOD\\, $DE) )\\, 0\\, 255) \\\\: x\\\\: 2 @}"
7208 Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7209 and the @option{fontsize} value are included in the @option{y} offset.
7211 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7212 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7217 For more information about libfreetype, check:
7218 @url{http://www.freetype.org/}.
7220 For more information about fontconfig, check:
7221 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7223 For more information about libfribidi, check:
7224 @url{http://fribidi.org/}.
7228 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7230 The filter accepts the following options:
7235 Set low and high threshold values used by the Canny thresholding
7238 The high threshold selects the "strong" edge pixels, which are then
7239 connected through 8-connectivity with the "weak" edge pixels selected
7240 by the low threshold.
7242 @var{low} and @var{high} threshold values must be chosen in the range
7243 [0,1], and @var{low} should be lesser or equal to @var{high}.
7245 Default value for @var{low} is @code{20/255}, and default value for @var{high}
7249 Define the drawing mode.
7253 Draw white/gray wires on black background.
7256 Mix the colors to create a paint/cartoon effect.
7259 Default value is @var{wires}.
7262 @subsection Examples
7266 Standard edge detection with custom values for the hysteresis thresholding:
7268 edgedetect=low=0.1:high=0.4
7272 Painting effect without thresholding:
7274 edgedetect=mode=colormix:high=0
7279 Set brightness, contrast, saturation and approximate gamma adjustment.
7281 The filter accepts the following options:
7285 Set the contrast expression. The value must be a float value in range
7286 @code{-2.0} to @code{2.0}. The default value is "1".
7289 Set the brightness expression. The value must be a float value in
7290 range @code{-1.0} to @code{1.0}. The default value is "0".
7293 Set the saturation expression. The value must be a float in
7294 range @code{0.0} to @code{3.0}. The default value is "1".
7297 Set the gamma expression. The value must be a float in range
7298 @code{0.1} to @code{10.0}. The default value is "1".
7301 Set the gamma expression for red. The value must be a float in
7302 range @code{0.1} to @code{10.0}. The default value is "1".
7305 Set the gamma expression for green. The value must be a float in range
7306 @code{0.1} to @code{10.0}. The default value is "1".
7309 Set the gamma expression for blue. The value must be a float in range
7310 @code{0.1} to @code{10.0}. The default value is "1".
7313 Set the gamma weight expression. It can be used to reduce the effect
7314 of a high gamma value on bright image areas, e.g. keep them from
7315 getting overamplified and just plain white. The value must be a float
7316 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7317 gamma correction all the way down while @code{1.0} leaves it at its
7318 full strength. Default is "1".
7321 Set when the expressions for brightness, contrast, saturation and
7322 gamma expressions are evaluated.
7324 It accepts the following values:
7327 only evaluate expressions once during the filter initialization or
7328 when a command is processed
7331 evaluate expressions for each incoming frame
7334 Default value is @samp{init}.
7337 The expressions accept the following parameters:
7340 frame count of the input frame starting from 0
7343 byte position of the corresponding packet in the input file, NAN if
7347 frame rate of the input video, NAN if the input frame rate is unknown
7350 timestamp expressed in seconds, NAN if the input timestamp is unknown
7353 @subsection Commands
7354 The filter supports the following commands:
7358 Set the contrast expression.
7361 Set the brightness expression.
7364 Set the saturation expression.
7367 Set the gamma expression.
7370 Set the gamma_r expression.
7373 Set gamma_g expression.
7376 Set gamma_b expression.
7379 Set gamma_weight expression.
7381 The command accepts the same syntax of the corresponding option.
7383 If the specified expression is not valid, it is kept at its current
7390 Apply erosion effect to the video.
7392 This filter replaces the pixel by the local(3x3) minimum.
7394 It accepts the following options:
7401 Limit the maximum change for each plane, default is 65535.
7402 If 0, plane will remain unchanged.
7405 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7408 Flags to local 3x3 coordinates maps like this:
7415 @section extractplanes
7417 Extract color channel components from input video stream into
7418 separate grayscale video streams.
7420 The filter accepts the following option:
7424 Set plane(s) to extract.
7426 Available values for planes are:
7437 Choosing planes not available in the input will result in an error.
7438 That means you cannot select @code{r}, @code{g}, @code{b} planes
7439 with @code{y}, @code{u}, @code{v} planes at same time.
7442 @subsection Examples
7446 Extract luma, u and v color channel component from input video frame
7447 into 3 grayscale outputs:
7449 ffmpeg -i video.avi -filter_complex 'extractplanes=y+u+v[y][u][v]' -map '[y]' y.avi -map '[u]' u.avi -map '[v]' v.avi
7455 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7457 For each input image, the filter will compute the optimal mapping from
7458 the input to the output given the codebook length, that is the number
7459 of distinct output colors.
7461 This filter accepts the following options.
7464 @item codebook_length, l
7465 Set codebook length. The value must be a positive integer, and
7466 represents the number of distinct output colors. Default value is 256.
7469 Set the maximum number of iterations to apply for computing the optimal
7470 mapping. The higher the value the better the result and the higher the
7471 computation time. Default value is 1.
7474 Set a random seed, must be an integer included between 0 and
7475 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7476 will try to use a good random seed on a best effort basis.
7479 Set pal8 output pixel format. This option does not work with codebook
7480 length greater than 256.
7485 Apply a fade-in/out effect to the input video.
7487 It accepts the following parameters:
7491 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7493 Default is @code{in}.
7495 @item start_frame, s
7496 Specify the number of the frame to start applying the fade
7497 effect at. Default is 0.
7500 The number of frames that the fade effect lasts. At the end of the
7501 fade-in effect, the output video will have the same intensity as the input video.
7502 At the end of the fade-out transition, the output video will be filled with the
7503 selected @option{color}.
7507 If set to 1, fade only alpha channel, if one exists on the input.
7510 @item start_time, st
7511 Specify the timestamp (in seconds) of the frame to start to apply the fade
7512 effect. If both start_frame and start_time are specified, the fade will start at
7513 whichever comes last. Default is 0.
7516 The number of seconds for which the fade effect has to last. At the end of the
7517 fade-in effect the output video will have the same intensity as the input video,
7518 at the end of the fade-out transition the output video will be filled with the
7519 selected @option{color}.
7520 If both duration and nb_frames are specified, duration is used. Default is 0
7521 (nb_frames is used by default).
7524 Specify the color of the fade. Default is "black".
7527 @subsection Examples
7531 Fade in the first 30 frames of video:
7536 The command above is equivalent to:
7542 Fade out the last 45 frames of a 200-frame video:
7545 fade=type=out:start_frame=155:nb_frames=45
7549 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7551 fade=in:0:25, fade=out:975:25
7555 Make the first 5 frames yellow, then fade in from frame 5-24:
7557 fade=in:5:20:color=yellow
7561 Fade in alpha over first 25 frames of video:
7563 fade=in:0:25:alpha=1
7567 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7569 fade=t=in:st=5.5:d=0.5
7575 Apply arbitrary expressions to samples in frequency domain
7579 Adjust the dc value (gain) of the luma plane of the image. The filter
7580 accepts an integer value in range @code{0} to @code{1000}. The default
7581 value is set to @code{0}.
7584 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7585 filter accepts an integer value in range @code{0} to @code{1000}. The
7586 default value is set to @code{0}.
7589 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7590 filter accepts an integer value in range @code{0} to @code{1000}. The
7591 default value is set to @code{0}.
7594 Set the frequency domain weight expression for the luma plane.
7597 Set the frequency domain weight expression for the 1st chroma plane.
7600 Set the frequency domain weight expression for the 2nd chroma plane.
7602 The filter accepts the following variables:
7605 The coordinates of the current sample.
7609 The width and height of the image.
7612 @subsection Examples
7618 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7624 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7630 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7636 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7643 Extract a single field from an interlaced image using stride
7644 arithmetic to avoid wasting CPU time. The output frames are marked as
7647 The filter accepts the following options:
7651 Specify whether to extract the top (if the value is @code{0} or
7652 @code{top}) or the bottom field (if the value is @code{1} or
7658 Create new frames by copying the top and bottom fields from surrounding frames
7659 supplied as numbers by the hint file.
7663 Set file containing hints: absolute/relative frame numbers.
7665 There must be one line for each frame in a clip. Each line must contain two
7666 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7667 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7668 is current frame number for @code{absolute} mode or out of [-1, 1] range
7669 for @code{relative} mode. First number tells from which frame to pick up top
7670 field and second number tells from which frame to pick up bottom field.
7672 If optionally followed by @code{+} output frame will be marked as interlaced,
7673 else if followed by @code{-} output frame will be marked as progressive, else
7674 it will be marked same as input frame.
7675 If line starts with @code{#} or @code{;} that line is skipped.
7678 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7681 Example of first several lines of @code{hint} file for @code{relative} mode:
7684 1,0 - # second frame, use third's frame top field and second's frame bottom field
7685 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7702 Field matching filter for inverse telecine. It is meant to reconstruct the
7703 progressive frames from a telecined stream. The filter does not drop duplicated
7704 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7705 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7707 The separation of the field matching and the decimation is notably motivated by
7708 the possibility of inserting a de-interlacing filter fallback between the two.
7709 If the source has mixed telecined and real interlaced content,
7710 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7711 But these remaining combed frames will be marked as interlaced, and thus can be
7712 de-interlaced by a later filter such as @ref{yadif} before decimation.
7714 In addition to the various configuration options, @code{fieldmatch} can take an
7715 optional second stream, activated through the @option{ppsrc} option. If
7716 enabled, the frames reconstruction will be based on the fields and frames from
7717 this second stream. This allows the first input to be pre-processed in order to
7718 help the various algorithms of the filter, while keeping the output lossless
7719 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7720 or brightness/contrast adjustments can help.
7722 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7723 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7724 which @code{fieldmatch} is based on. While the semantic and usage are very
7725 close, some behaviour and options names can differ.
7727 The @ref{decimate} filter currently only works for constant frame rate input.
7728 If your input has mixed telecined (30fps) and progressive content with a lower
7729 framerate like 24fps use the following filterchain to produce the necessary cfr
7730 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7732 The filter accepts the following options:
7736 Specify the assumed field order of the input stream. Available values are:
7740 Auto detect parity (use FFmpeg's internal parity value).
7742 Assume bottom field first.
7744 Assume top field first.
7747 Note that it is sometimes recommended not to trust the parity announced by the
7750 Default value is @var{auto}.
7753 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7754 sense that it won't risk creating jerkiness due to duplicate frames when
7755 possible, but if there are bad edits or blended fields it will end up
7756 outputting combed frames when a good match might actually exist. On the other
7757 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7758 but will almost always find a good frame if there is one. The other values are
7759 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7760 jerkiness and creating duplicate frames versus finding good matches in sections
7761 with bad edits, orphaned fields, blended fields, etc.
7763 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7765 Available values are:
7769 2-way matching (p/c)
7771 2-way matching, and trying 3rd match if still combed (p/c + n)
7773 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7775 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7776 still combed (p/c + n + u/b)
7778 3-way matching (p/c/n)
7780 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7781 detected as combed (p/c/n + u/b)
7784 The parenthesis at the end indicate the matches that would be used for that
7785 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7788 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7791 Default value is @var{pc_n}.
7794 Mark the main input stream as a pre-processed input, and enable the secondary
7795 input stream as the clean source to pick the fields from. See the filter
7796 introduction for more details. It is similar to the @option{clip2} feature from
7799 Default value is @code{0} (disabled).
7802 Set the field to match from. It is recommended to set this to the same value as
7803 @option{order} unless you experience matching failures with that setting. In
7804 certain circumstances changing the field that is used to match from can have a
7805 large impact on matching performance. Available values are:
7809 Automatic (same value as @option{order}).
7811 Match from the bottom field.
7813 Match from the top field.
7816 Default value is @var{auto}.
7819 Set whether or not chroma is included during the match comparisons. In most
7820 cases it is recommended to leave this enabled. You should set this to @code{0}
7821 only if your clip has bad chroma problems such as heavy rainbowing or other
7822 artifacts. Setting this to @code{0} could also be used to speed things up at
7823 the cost of some accuracy.
7825 Default value is @code{1}.
7829 These define an exclusion band which excludes the lines between @option{y0} and
7830 @option{y1} from being included in the field matching decision. An exclusion
7831 band can be used to ignore subtitles, a logo, or other things that may
7832 interfere with the matching. @option{y0} sets the starting scan line and
7833 @option{y1} sets the ending line; all lines in between @option{y0} and
7834 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7835 @option{y0} and @option{y1} to the same value will disable the feature.
7836 @option{y0} and @option{y1} defaults to @code{0}.
7839 Set the scene change detection threshold as a percentage of maximum change on
7840 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7841 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7842 @option{scthresh} is @code{[0.0, 100.0]}.
7844 Default value is @code{12.0}.
7847 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7848 account the combed scores of matches when deciding what match to use as the
7849 final match. Available values are:
7853 No final matching based on combed scores.
7855 Combed scores are only used when a scene change is detected.
7857 Use combed scores all the time.
7860 Default is @var{sc}.
7863 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7864 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7865 Available values are:
7869 No forced calculation.
7871 Force p/c/n calculations.
7873 Force p/c/n/u/b calculations.
7876 Default value is @var{none}.
7879 This is the area combing threshold used for combed frame detection. This
7880 essentially controls how "strong" or "visible" combing must be to be detected.
7881 Larger values mean combing must be more visible and smaller values mean combing
7882 can be less visible or strong and still be detected. Valid settings are from
7883 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7884 be detected as combed). This is basically a pixel difference value. A good
7885 range is @code{[8, 12]}.
7887 Default value is @code{9}.
7890 Sets whether or not chroma is considered in the combed frame decision. Only
7891 disable this if your source has chroma problems (rainbowing, etc.) that are
7892 causing problems for the combed frame detection with chroma enabled. Actually,
7893 using @option{chroma}=@var{0} is usually more reliable, except for the case
7894 where there is chroma only combing in the source.
7896 Default value is @code{0}.
7900 Respectively set the x-axis and y-axis size of the window used during combed
7901 frame detection. This has to do with the size of the area in which
7902 @option{combpel} pixels are required to be detected as combed for a frame to be
7903 declared combed. See the @option{combpel} parameter description for more info.
7904 Possible values are any number that is a power of 2 starting at 4 and going up
7907 Default value is @code{16}.
7910 The number of combed pixels inside any of the @option{blocky} by
7911 @option{blockx} size blocks on the frame for the frame to be detected as
7912 combed. While @option{cthresh} controls how "visible" the combing must be, this
7913 setting controls "how much" combing there must be in any localized area (a
7914 window defined by the @option{blockx} and @option{blocky} settings) on the
7915 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7916 which point no frames will ever be detected as combed). This setting is known
7917 as @option{MI} in TFM/VFM vocabulary.
7919 Default value is @code{80}.
7922 @anchor{p/c/n/u/b meaning}
7923 @subsection p/c/n/u/b meaning
7925 @subsubsection p/c/n
7927 We assume the following telecined stream:
7930 Top fields: 1 2 2 3 4
7931 Bottom fields: 1 2 3 4 4
7934 The numbers correspond to the progressive frame the fields relate to. Here, the
7935 first two frames are progressive, the 3rd and 4th are combed, and so on.
7937 When @code{fieldmatch} is configured to run a matching from bottom
7938 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7943 B 1 2 3 4 4 <-- matching reference
7952 As a result of the field matching, we can see that some frames get duplicated.
7953 To perform a complete inverse telecine, you need to rely on a decimation filter
7954 after this operation. See for instance the @ref{decimate} filter.
7956 The same operation now matching from top fields (@option{field}=@var{top})
7961 T 1 2 2 3 4 <-- matching reference
7971 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7972 basically, they refer to the frame and field of the opposite parity:
7975 @item @var{p} matches the field of the opposite parity in the previous frame
7976 @item @var{c} matches the field of the opposite parity in the current frame
7977 @item @var{n} matches the field of the opposite parity in the next frame
7982 The @var{u} and @var{b} matching are a bit special in the sense that they match
7983 from the opposite parity flag. In the following examples, we assume that we are
7984 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7985 'x' is placed above and below each matched fields.
7987 With bottom matching (@option{field}=@var{bottom}):
7992 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7993 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
8001 With top matching (@option{field}=@var{top}):
8006 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
8007 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
8015 @subsection Examples
8017 Simple IVTC of a top field first telecined stream:
8019 fieldmatch=order=tff:combmatch=none, decimate
8022 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
8024 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
8029 Transform the field order of the input video.
8031 It accepts the following parameters:
8036 The output field order. Valid values are @var{tff} for top field first or @var{bff}
8037 for bottom field first.
8040 The default value is @samp{tff}.
8042 The transformation is done by shifting the picture content up or down
8043 by one line, and filling the remaining line with appropriate picture content.
8044 This method is consistent with most broadcast field order converters.
8046 If the input video is not flagged as being interlaced, or it is already
8047 flagged as being of the required output field order, then this filter does
8048 not alter the incoming video.
8050 It is very useful when converting to or from PAL DV material,
8051 which is bottom field first.
8055 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
8058 @section fifo, afifo
8060 Buffer input images and send them when they are requested.
8062 It is mainly useful when auto-inserted by the libavfilter
8065 It does not take parameters.
8069 Find a rectangular object
8071 It accepts the following options:
8075 Filepath of the object image, needs to be in gray8.
8078 Detection threshold, default is 0.5.
8081 Number of mipmaps, default is 3.
8083 @item xmin, ymin, xmax, ymax
8084 Specifies the rectangle in which to search.
8087 @subsection Examples
8091 Generate a representative palette of a given video using @command{ffmpeg}:
8093 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8099 Cover a rectangular object
8101 It accepts the following options:
8105 Filepath of the optional cover image, needs to be in yuv420.
8110 It accepts the following values:
8113 cover it by the supplied image
8115 cover it by interpolating the surrounding pixels
8118 Default value is @var{blur}.
8121 @subsection Examples
8125 Generate a representative palette of a given video using @command{ffmpeg}:
8127 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8134 Convert the input video to one of the specified pixel formats.
8135 Libavfilter will try to pick one that is suitable as input to
8138 It accepts the following parameters:
8142 A '|'-separated list of pixel format names, such as
8143 "pix_fmts=yuv420p|monow|rgb24".
8147 @subsection Examples
8151 Convert the input video to the @var{yuv420p} format
8153 format=pix_fmts=yuv420p
8156 Convert the input video to any of the formats in the list
8158 format=pix_fmts=yuv420p|yuv444p|yuv410p
8165 Convert the video to specified constant frame rate by duplicating or dropping
8166 frames as necessary.
8168 It accepts the following parameters:
8172 The desired output frame rate. The default is @code{25}.
8177 Possible values are:
8180 zero round towards 0
8184 round towards -infinity
8186 round towards +infinity
8190 The default is @code{near}.
8193 Assume the first PTS should be the given value, in seconds. This allows for
8194 padding/trimming at the start of stream. By default, no assumption is made
8195 about the first frame's expected PTS, so no padding or trimming is done.
8196 For example, this could be set to 0 to pad the beginning with duplicates of
8197 the first frame if a video stream starts after the audio stream or to trim any
8198 frames with a negative PTS.
8202 Alternatively, the options can be specified as a flat string:
8203 @var{fps}[:@var{round}].
8205 See also the @ref{setpts} filter.
8207 @subsection Examples
8211 A typical usage in order to set the fps to 25:
8217 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8219 fps=fps=film:round=near
8225 Pack two different video streams into a stereoscopic video, setting proper
8226 metadata on supported codecs. The two views should have the same size and
8227 framerate and processing will stop when the shorter video ends. Please note
8228 that you may conveniently adjust view properties with the @ref{scale} and
8231 It accepts the following parameters:
8235 The desired packing format. Supported values are:
8240 The views are next to each other (default).
8243 The views are on top of each other.
8246 The views are packed by line.
8249 The views are packed by column.
8252 The views are temporally interleaved.
8261 # Convert left and right views into a frame-sequential video
8262 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8264 # Convert views into a side-by-side video with the same output resolution as the input
8265 ffmpeg -i LEFT -i RIGHT -filter_complex [0:v]scale=w=iw/2[left],[1:v]scale=w=iw/2[right],[left][right]framepack=sbs OUTPUT
8270 Change the frame rate by interpolating new video output frames from the source
8273 This filter is not designed to function correctly with interlaced media. If
8274 you wish to change the frame rate of interlaced media then you are required
8275 to deinterlace before this filter and re-interlace after this filter.
8277 A description of the accepted options follows.
8281 Specify the output frames per second. This option can also be specified
8282 as a value alone. The default is @code{50}.
8285 Specify the start of a range where the output frame will be created as a
8286 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8287 the default is @code{15}.
8290 Specify the end of a range where the output frame will be created as a
8291 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8292 the default is @code{240}.
8295 Specify the level at which a scene change is detected as a value between
8296 0 and 100 to indicate a new scene; a low value reflects a low
8297 probability for the current frame to introduce a new scene, while a higher
8298 value means the current frame is more likely to be one.
8299 The default is @code{7}.
8302 Specify flags influencing the filter process.
8304 Available value for @var{flags} is:
8307 @item scene_change_detect, scd
8308 Enable scene change detection using the value of the option @var{scene}.
8309 This flag is enabled by default.
8315 Select one frame every N-th frame.
8317 This filter accepts the following option:
8320 Select frame after every @code{step} frames.
8321 Allowed values are positive integers higher than 0. Default value is @code{1}.
8327 Apply a frei0r effect to the input video.
8329 To enable the compilation of this filter, you need to install the frei0r
8330 header and configure FFmpeg with @code{--enable-frei0r}.
8332 It accepts the following parameters:
8337 The name of the frei0r effect to load. If the environment variable
8338 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8339 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8340 Otherwise, the standard frei0r paths are searched, in this order:
8341 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8342 @file{/usr/lib/frei0r-1/}.
8345 A '|'-separated list of parameters to pass to the frei0r effect.
8349 A frei0r effect parameter can be a boolean (its value is either
8350 "y" or "n"), a double, a color (specified as
8351 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8352 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8353 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8354 @var{X} and @var{Y} are floating point numbers) and/or a string.
8356 The number and types of parameters depend on the loaded effect. If an
8357 effect parameter is not specified, the default value is set.
8359 @subsection Examples
8363 Apply the distort0r effect, setting the first two double parameters:
8365 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8369 Apply the colordistance effect, taking a color as the first parameter:
8371 frei0r=colordistance:0.2/0.3/0.4
8372 frei0r=colordistance:violet
8373 frei0r=colordistance:0x112233
8377 Apply the perspective effect, specifying the top left and top right image
8380 frei0r=perspective:0.2/0.2|0.8/0.2
8384 For more information, see
8385 @url{http://frei0r.dyne.org}
8389 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8391 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8392 processing filter, one of them is performed once per block, not per pixel.
8393 This allows for much higher speed.
8395 The filter accepts the following options:
8399 Set quality. This option defines the number of levels for averaging. It accepts
8400 an integer in the range 4-5. Default value is @code{4}.
8403 Force a constant quantization parameter. It accepts an integer in range 0-63.
8404 If not set, the filter will use the QP from the video stream (if available).
8407 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8408 more details but also more artifacts, while higher values make the image smoother
8409 but also blurrier. Default value is @code{0} − PSNR optimal.
8412 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8413 option may cause flicker since the B-Frames have often larger QP. Default is
8414 @code{0} (not enabled).
8420 Apply Gaussian blur filter.
8422 The filter accepts the following options:
8426 Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8429 Set number of steps for Gaussian approximation. Defauls is @code{1}.
8432 Set which planes to filter. By default all planes are filtered.
8435 Set vertical sigma, if negative it will be same as @code{sigma}.
8436 Default is @code{-1}.
8441 The filter accepts the following options:
8445 Set the luminance expression.
8447 Set the chrominance blue expression.
8449 Set the chrominance red expression.
8451 Set the alpha expression.
8453 Set the red expression.
8455 Set the green expression.
8457 Set the blue expression.
8460 The colorspace is selected according to the specified options. If one
8461 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8462 options is specified, the filter will automatically select a YCbCr
8463 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8464 @option{blue_expr} options is specified, it will select an RGB
8467 If one of the chrominance expression is not defined, it falls back on the other
8468 one. If no alpha expression is specified it will evaluate to opaque value.
8469 If none of chrominance expressions are specified, they will evaluate
8470 to the luminance expression.
8472 The expressions can use the following variables and functions:
8476 The sequential number of the filtered frame, starting from @code{0}.
8480 The coordinates of the current sample.
8484 The width and height of the image.
8488 Width and height scale depending on the currently filtered plane. It is the
8489 ratio between the corresponding luma plane number of pixels and the current
8490 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8491 @code{0.5,0.5} for chroma planes.
8494 Time of the current frame, expressed in seconds.
8497 Return the value of the pixel at location (@var{x},@var{y}) of the current
8501 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8505 Return the value of the pixel at location (@var{x},@var{y}) of the
8506 blue-difference chroma plane. Return 0 if there is no such plane.
8509 Return the value of the pixel at location (@var{x},@var{y}) of the
8510 red-difference chroma plane. Return 0 if there is no such plane.
8515 Return the value of the pixel at location (@var{x},@var{y}) of the
8516 red/green/blue component. Return 0 if there is no such component.
8519 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8520 plane. Return 0 if there is no such plane.
8523 For functions, if @var{x} and @var{y} are outside the area, the value will be
8524 automatically clipped to the closer edge.
8526 @subsection Examples
8530 Flip the image horizontally:
8536 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8537 wavelength of 100 pixels:
8539 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8543 Generate a fancy enigmatic moving light:
8545 nullsrc=s=256x256,geq=random(1)/hypot(X-cos(N*0.07)*W/2-W/2\,Y-sin(N*0.09)*H/2-H/2)^2*1000000*sin(N*0.02):128:128
8549 Generate a quick emboss effect:
8551 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8555 Modify RGB components depending on pixel position:
8557 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8561 Create a radial gradient that is the same size as the input (also see
8562 the @ref{vignette} filter):
8564 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8570 Fix the banding artifacts that are sometimes introduced into nearly flat
8571 regions by truncation to 8-bit color depth.
8572 Interpolate the gradients that should go where the bands are, and
8575 It is designed for playback only. Do not use it prior to
8576 lossy compression, because compression tends to lose the dither and
8577 bring back the bands.
8579 It accepts the following parameters:
8584 The maximum amount by which the filter will change any one pixel. This is also
8585 the threshold for detecting nearly flat regions. Acceptable values range from
8586 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8590 The neighborhood to fit the gradient to. A larger radius makes for smoother
8591 gradients, but also prevents the filter from modifying the pixels near detailed
8592 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8593 values will be clipped to the valid range.
8597 Alternatively, the options can be specified as a flat string:
8598 @var{strength}[:@var{radius}]
8600 @subsection Examples
8604 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8610 Specify radius, omitting the strength (which will fall-back to the default
8621 Apply a Hald CLUT to a video stream.
8623 First input is the video stream to process, and second one is the Hald CLUT.
8624 The Hald CLUT input can be a simple picture or a complete video stream.
8626 The filter accepts the following options:
8630 Force termination when the shortest input terminates. Default is @code{0}.
8632 Continue applying the last CLUT after the end of the stream. A value of
8633 @code{0} disable the filter after the last frame of the CLUT is reached.
8634 Default is @code{1}.
8637 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8638 filters share the same internals).
8640 More information about the Hald CLUT can be found on Eskil Steenberg's website
8641 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8643 @subsection Workflow examples
8645 @subsubsection Hald CLUT video stream
8647 Generate an identity Hald CLUT stream altered with various effects:
8649 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
8652 Note: make sure you use a lossless codec.
8654 Then use it with @code{haldclut} to apply it on some random stream:
8656 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8659 The Hald CLUT will be applied to the 10 first seconds (duration of
8660 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8661 to the remaining frames of the @code{mandelbrot} stream.
8663 @subsubsection Hald CLUT with preview
8665 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8666 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8667 biggest possible square starting at the top left of the picture. The remaining
8668 padding pixels (bottom or right) will be ignored. This area can be used to add
8669 a preview of the Hald CLUT.
8671 Typically, the following generated Hald CLUT will be supported by the
8672 @code{haldclut} filter:
8675 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8676 pad=iw+320 [padded_clut];
8677 smptebars=s=320x256, split [a][b];
8678 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8679 [main][b] overlay=W-320" -frames:v 1 clut.png
8682 It contains the original and a preview of the effect of the CLUT: SMPTE color
8683 bars are displayed on the right-top, and below the same color bars processed by
8686 Then, the effect of this Hald CLUT can be visualized with:
8688 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8693 Flip the input video horizontally.
8695 For example, to horizontally flip the input video with @command{ffmpeg}:
8697 ffmpeg -i in.avi -vf "hflip" out.avi
8701 This filter applies a global color histogram equalization on a
8704 It can be used to correct video that has a compressed range of pixel
8705 intensities. The filter redistributes the pixel intensities to
8706 equalize their distribution across the intensity range. It may be
8707 viewed as an "automatically adjusting contrast filter". This filter is
8708 useful only for correcting degraded or poorly captured source
8711 The filter accepts the following options:
8715 Determine the amount of equalization to be applied. As the strength
8716 is reduced, the distribution of pixel intensities more-and-more
8717 approaches that of the input frame. The value must be a float number
8718 in the range [0,1] and defaults to 0.200.
8721 Set the maximum intensity that can generated and scale the output
8722 values appropriately. The strength should be set as desired and then
8723 the intensity can be limited if needed to avoid washing-out. The value
8724 must be a float number in the range [0,1] and defaults to 0.210.
8727 Set the antibanding level. If enabled the filter will randomly vary
8728 the luminance of output pixels by a small amount to avoid banding of
8729 the histogram. Possible values are @code{none}, @code{weak} or
8730 @code{strong}. It defaults to @code{none}.
8735 Compute and draw a color distribution histogram for the input video.
8737 The computed histogram is a representation of the color component
8738 distribution in an image.
8740 Standard histogram displays the color components distribution in an image.
8741 Displays color graph for each color component. Shows distribution of
8742 the Y, U, V, A or R, G, B components, depending on input format, in the
8743 current frame. Below each graph a color component scale meter is shown.
8745 The filter accepts the following options:
8749 Set height of level. Default value is @code{200}.
8750 Allowed range is [50, 2048].
8753 Set height of color scale. Default value is @code{12}.
8754 Allowed range is [0, 40].
8758 It accepts the following values:
8761 Per color component graphs are placed below each other.
8764 Presents information identical to that in the @code{parade}, except
8765 that the graphs representing color components are superimposed directly
8768 Default is @code{parade}.
8771 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8772 Default is @code{linear}.
8775 Set what color components to display.
8776 Default is @code{7}.
8779 Set foreground opacity. Default is @code{0.7}.
8782 Set background opacity. Default is @code{0.5}.
8785 @subsection Examples
8790 Calculate and draw histogram:
8792 ffplay -i input -vf histogram
8800 This is a high precision/quality 3d denoise filter. It aims to reduce
8801 image noise, producing smooth images and making still images really
8802 still. It should enhance compressibility.
8804 It accepts the following optional parameters:
8808 A non-negative floating point number which specifies spatial luma strength.
8811 @item chroma_spatial
8812 A non-negative floating point number which specifies spatial chroma strength.
8813 It defaults to 3.0*@var{luma_spatial}/4.0.
8816 A floating point number which specifies luma temporal strength. It defaults to
8817 6.0*@var{luma_spatial}/4.0.
8820 A floating point number which specifies chroma temporal strength. It defaults to
8821 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8824 @anchor{hwupload_cuda}
8825 @section hwupload_cuda
8827 Upload system memory frames to a CUDA device.
8829 It accepts the following optional parameters:
8833 The number of the CUDA device to use
8838 Apply a high-quality magnification filter designed for pixel art. This filter
8839 was originally created by Maxim Stepin.
8841 It accepts the following option:
8845 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8846 @code{hq3x} and @code{4} for @code{hq4x}.
8847 Default is @code{3}.
8851 Stack input videos horizontally.
8853 All streams must be of same pixel format and of same height.
8855 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8856 to create same output.
8858 The filter accept the following option:
8862 Set number of input streams. Default is 2.
8865 If set to 1, force the output to terminate when the shortest input
8866 terminates. Default value is 0.
8871 Modify the hue and/or the saturation of the input.
8873 It accepts the following parameters:
8877 Specify the hue angle as a number of degrees. It accepts an expression,
8878 and defaults to "0".
8881 Specify the saturation in the [-10,10] range. It accepts an expression and
8885 Specify the hue angle as a number of radians. It accepts an
8886 expression, and defaults to "0".
8889 Specify the brightness in the [-10,10] range. It accepts an expression and
8893 @option{h} and @option{H} are mutually exclusive, and can't be
8894 specified at the same time.
8896 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8897 expressions containing the following constants:
8901 frame count of the input frame starting from 0
8904 presentation timestamp of the input frame expressed in time base units
8907 frame rate of the input video, NAN if the input frame rate is unknown
8910 timestamp expressed in seconds, NAN if the input timestamp is unknown
8913 time base of the input video
8916 @subsection Examples
8920 Set the hue to 90 degrees and the saturation to 1.0:
8926 Same command but expressing the hue in radians:
8932 Rotate hue and make the saturation swing between 0
8933 and 2 over a period of 1 second:
8935 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8939 Apply a 3 seconds saturation fade-in effect starting at 0:
8944 The general fade-in expression can be written as:
8946 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8950 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8952 hue="s=max(0\, min(1\, (8-t)/3))"
8955 The general fade-out expression can be written as:
8957 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8962 @subsection Commands
8964 This filter supports the following commands:
8970 Modify the hue and/or the saturation and/or brightness of the input video.
8971 The command accepts the same syntax of the corresponding option.
8973 If the specified expression is not valid, it is kept at its current
8979 Grow first stream into second stream by connecting components.
8980 This makes it possible to build more robust edge masks.
8982 This filter accepts the following options:
8986 Set which planes will be processed as bitmap, unprocessed planes will be
8987 copied from first stream.
8988 By default value 0xf, all planes will be processed.
8991 Set threshold which is used in filtering. If pixel component value is higher than
8992 this value filter algorithm for connecting components is activated.
8993 By default value is 0.
8998 Detect video interlacing type.
9000 This filter tries to detect if the input frames are interlaced, progressive,
9001 top or bottom field first. It will also try to detect fields that are
9002 repeated between adjacent frames (a sign of telecine).
9004 Single frame detection considers only immediately adjacent frames when classifying each frame.
9005 Multiple frame detection incorporates the classification history of previous frames.
9007 The filter will log these metadata values:
9010 @item single.current_frame
9011 Detected type of current frame using single-frame detection. One of:
9012 ``tff'' (top field first), ``bff'' (bottom field first),
9013 ``progressive'', or ``undetermined''
9016 Cumulative number of frames detected as top field first using single-frame detection.
9019 Cumulative number of frames detected as top field first using multiple-frame detection.
9022 Cumulative number of frames detected as bottom field first using single-frame detection.
9024 @item multiple.current_frame
9025 Detected type of current frame using multiple-frame detection. One of:
9026 ``tff'' (top field first), ``bff'' (bottom field first),
9027 ``progressive'', or ``undetermined''
9030 Cumulative number of frames detected as bottom field first using multiple-frame detection.
9032 @item single.progressive
9033 Cumulative number of frames detected as progressive using single-frame detection.
9035 @item multiple.progressive
9036 Cumulative number of frames detected as progressive using multiple-frame detection.
9038 @item single.undetermined
9039 Cumulative number of frames that could not be classified using single-frame detection.
9041 @item multiple.undetermined
9042 Cumulative number of frames that could not be classified using multiple-frame detection.
9044 @item repeated.current_frame
9045 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
9047 @item repeated.neither
9048 Cumulative number of frames with no repeated field.
9051 Cumulative number of frames with the top field repeated from the previous frame's top field.
9053 @item repeated.bottom
9054 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
9057 The filter accepts the following options:
9061 Set interlacing threshold.
9063 Set progressive threshold.
9065 Threshold for repeated field detection.
9067 Number of frames after which a given frame's contribution to the
9068 statistics is halved (i.e., it contributes only 0.5 to its
9069 classification). The default of 0 means that all frames seen are given
9070 full weight of 1.0 forever.
9071 @item analyze_interlaced_flag
9072 When this is not 0 then idet will use the specified number of frames to determine
9073 if the interlaced flag is accurate, it will not count undetermined frames.
9074 If the flag is found to be accurate it will be used without any further
9075 computations, if it is found to be inaccurate it will be cleared without any
9076 further computations. This allows inserting the idet filter as a low computational
9077 method to clean up the interlaced flag
9082 Deinterleave or interleave fields.
9084 This filter allows one to process interlaced images fields without
9085 deinterlacing them. Deinterleaving splits the input frame into 2
9086 fields (so called half pictures). Odd lines are moved to the top
9087 half of the output image, even lines to the bottom half.
9088 You can process (filter) them independently and then re-interleave them.
9090 The filter accepts the following options:
9094 @item chroma_mode, c
9096 Available values for @var{luma_mode}, @var{chroma_mode} and
9097 @var{alpha_mode} are:
9103 @item deinterleave, d
9104 Deinterleave fields, placing one above the other.
9107 Interleave fields. Reverse the effect of deinterleaving.
9109 Default value is @code{none}.
9112 @item chroma_swap, cs
9113 @item alpha_swap, as
9114 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9119 Apply inflate effect to the video.
9121 This filter replaces the pixel by the local(3x3) average by taking into account
9122 only values higher than the pixel.
9124 It accepts the following options:
9131 Limit the maximum change for each plane, default is 65535.
9132 If 0, plane will remain unchanged.
9137 Simple interlacing filter from progressive contents. This interleaves upper (or
9138 lower) lines from odd frames with lower (or upper) lines from even frames,
9139 halving the frame rate and preserving image height.
9142 Original Original New Frame
9143 Frame 'j' Frame 'j+1' (tff)
9144 ========== =========== ==================
9145 Line 0 --------------------> Frame 'j' Line 0
9146 Line 1 Line 1 ----> Frame 'j+1' Line 1
9147 Line 2 ---------------------> Frame 'j' Line 2
9148 Line 3 Line 3 ----> Frame 'j+1' Line 3
9150 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9153 It accepts the following optional parameters:
9157 This determines whether the interlaced frame is taken from the even
9158 (tff - default) or odd (bff) lines of the progressive frame.
9161 Enable (default) or disable the vertical lowpass filter to avoid twitter
9162 interlacing and reduce moire patterns.
9167 Deinterlace input video by applying Donald Graft's adaptive kernel
9168 deinterling. Work on interlaced parts of a video to produce
9171 The description of the accepted parameters follows.
9175 Set the threshold which affects the filter's tolerance when
9176 determining if a pixel line must be processed. It must be an integer
9177 in the range [0,255] and defaults to 10. A value of 0 will result in
9178 applying the process on every pixels.
9181 Paint pixels exceeding the threshold value to white if set to 1.
9185 Set the fields order. Swap fields if set to 1, leave fields alone if
9189 Enable additional sharpening if set to 1. Default is 0.
9192 Enable twoway sharpening if set to 1. Default is 0.
9195 @subsection Examples
9199 Apply default values:
9201 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9205 Enable additional sharpening:
9211 Paint processed pixels in white:
9217 @section lenscorrection
9219 Correct radial lens distortion
9221 This filter can be used to correct for radial distortion as can result from the use
9222 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9223 one can use tools available for example as part of opencv or simply trial-and-error.
9224 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9225 and extract the k1 and k2 coefficients from the resulting matrix.
9227 Note that effectively the same filter is available in the open-source tools Krita and
9228 Digikam from the KDE project.
9230 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9231 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9232 brightness distribution, so you may want to use both filters together in certain
9233 cases, though you will have to take care of ordering, i.e. whether vignetting should
9234 be applied before or after lens correction.
9238 The filter accepts the following options:
9242 Relative x-coordinate of the focal point of the image, and thereby the center of the
9243 distortion. This value has a range [0,1] and is expressed as fractions of the image
9246 Relative y-coordinate of the focal point of the image, and thereby the center of the
9247 distortion. This value has a range [0,1] and is expressed as fractions of the image
9250 Coefficient of the quadratic correction term. 0.5 means no correction.
9252 Coefficient of the double quadratic correction term. 0.5 means no correction.
9255 The formula that generates the correction is:
9257 @var{r_src} = @var{r_tgt} * (1 + @var{k1} * (@var{r_tgt} / @var{r_0})^2 + @var{k2} * (@var{r_tgt} / @var{r_0})^4)
9259 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9260 distances from the focal point in the source and target images, respectively.
9266 The filter accepts the following options:
9270 Set the number of loops.
9273 Set maximal size in number of frames.
9276 Set first frame of loop.
9282 Apply a 3D LUT to an input video.
9284 The filter accepts the following options:
9288 Set the 3D LUT file name.
9290 Currently supported formats:
9302 Select interpolation mode.
9304 Available values are:
9308 Use values from the nearest defined point.
9310 Interpolate values using the 8 points defining a cube.
9312 Interpolate values using a tetrahedron.
9318 Turn certain luma values into transparency.
9320 The filter accepts the following options:
9324 Set the luma which will be used as base for transparency.
9325 Default value is @code{0}.
9328 Set the range of luma values to be keyed out.
9329 Default value is @code{0}.
9332 Set the range of softness. Default value is @code{0}.
9333 Use this to control gradual transition from zero to full transparency.
9336 @section lut, lutrgb, lutyuv
9338 Compute a look-up table for binding each pixel component input value
9339 to an output value, and apply it to the input video.
9341 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9342 to an RGB input video.
9344 These filters accept the following parameters:
9347 set first pixel component expression
9349 set second pixel component expression
9351 set third pixel component expression
9353 set fourth pixel component expression, corresponds to the alpha component
9356 set red component expression
9358 set green component expression
9360 set blue component expression
9362 alpha component expression
9365 set Y/luminance component expression
9367 set U/Cb component expression
9369 set V/Cr component expression
9372 Each of them specifies the expression to use for computing the lookup table for
9373 the corresponding pixel component values.
9375 The exact component associated to each of the @var{c*} options depends on the
9378 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9379 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9381 The expressions can contain the following constants and functions:
9386 The input width and height.
9389 The input value for the pixel component.
9392 The input value, clipped to the @var{minval}-@var{maxval} range.
9395 The maximum value for the pixel component.
9398 The minimum value for the pixel component.
9401 The negated value for the pixel component value, clipped to the
9402 @var{minval}-@var{maxval} range; it corresponds to the expression
9403 "maxval-clipval+minval".
9406 The computed value in @var{val}, clipped to the
9407 @var{minval}-@var{maxval} range.
9409 @item gammaval(gamma)
9410 The computed gamma correction value of the pixel component value,
9411 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9413 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9417 All expressions default to "val".
9419 @subsection Examples
9425 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9426 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9429 The above is the same as:
9431 lutrgb="r=negval:g=negval:b=negval"
9432 lutyuv="y=negval:u=negval:v=negval"
9442 Remove chroma components, turning the video into a graytone image:
9444 lutyuv="u=128:v=128"
9448 Apply a luma burning effect:
9454 Remove green and blue components:
9460 Set a constant alpha channel value on input:
9462 format=rgba,lutrgb=a="maxval-minval/2"
9466 Correct luminance gamma by a factor of 0.5:
9468 lutyuv=y=gammaval(0.5)
9472 Discard least significant bits of luma:
9474 lutyuv=y='bitand(val, 128+64+32)'
9478 Technicolor like effect:
9480 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9486 Compute and apply a lookup table from two video inputs.
9488 This filter accepts the following parameters:
9491 set first pixel component expression
9493 set second pixel component expression
9495 set third pixel component expression
9497 set fourth pixel component expression, corresponds to the alpha component
9500 Each of them specifies the expression to use for computing the lookup table for
9501 the corresponding pixel component values.
9503 The exact component associated to each of the @var{c*} options depends on the
9506 The expressions can contain the following constants:
9511 The input width and height.
9514 The first input value for the pixel component.
9517 The second input value for the pixel component.
9520 The first input video bit depth.
9523 The second input video bit depth.
9526 All expressions default to "x".
9528 @subsection Examples
9532 Highlight differences between two RGB video streams:
9534 lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1)'
9538 Highlight differences between two YUV video streams:
9540 lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1)'
9544 @section maskedclamp
9546 Clamp the first input stream with the second input and third input stream.
9548 Returns the value of first stream to be between second input
9549 stream - @code{undershoot} and third input stream + @code{overshoot}.
9551 This filter accepts the following options:
9554 Default value is @code{0}.
9557 Default value is @code{0}.
9560 Set which planes will be processed as bitmap, unprocessed planes will be
9561 copied from first stream.
9562 By default value 0xf, all planes will be processed.
9565 @section maskedmerge
9567 Merge the first input stream with the second input stream using per pixel
9568 weights in the third input stream.
9570 A value of 0 in the third stream pixel component means that pixel component
9571 from first stream is returned unchanged, while maximum value (eg. 255 for
9572 8-bit videos) means that pixel component from second stream is returned
9573 unchanged. Intermediate values define the amount of merging between both
9574 input stream's pixel components.
9576 This filter accepts the following options:
9579 Set which planes will be processed as bitmap, unprocessed planes will be
9580 copied from first stream.
9581 By default value 0xf, all planes will be processed.
9586 Apply motion-compensation deinterlacing.
9588 It needs one field per frame as input and must thus be used together
9589 with yadif=1/3 or equivalent.
9591 This filter accepts the following options:
9594 Set the deinterlacing mode.
9596 It accepts one of the following values:
9601 use iterative motion estimation
9603 like @samp{slow}, but use multiple reference frames.
9605 Default value is @samp{fast}.
9608 Set the picture field parity assumed for the input video. It must be
9609 one of the following values:
9613 assume top field first
9615 assume bottom field first
9618 Default value is @samp{bff}.
9621 Set per-block quantization parameter (QP) used by the internal
9624 Higher values should result in a smoother motion vector field but less
9625 optimal individual vectors. Default value is 1.
9628 @section mergeplanes
9630 Merge color channel components from several video streams.
9632 The filter accepts up to 4 input streams, and merge selected input
9633 planes to the output video.
9635 This filter accepts the following options:
9638 Set input to output plane mapping. Default is @code{0}.
9640 The mappings is specified as a bitmap. It should be specified as a
9641 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9642 mapping for the first plane of the output stream. 'A' sets the number of
9643 the input stream to use (from 0 to 3), and 'a' the plane number of the
9644 corresponding input to use (from 0 to 3). The rest of the mappings is
9645 similar, 'Bb' describes the mapping for the output stream second
9646 plane, 'Cc' describes the mapping for the output stream third plane and
9647 'Dd' describes the mapping for the output stream fourth plane.
9650 Set output pixel format. Default is @code{yuva444p}.
9653 @subsection Examples
9657 Merge three gray video streams of same width and height into single video stream:
9659 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9663 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9665 [a0][a1]mergeplanes=0x00010210:yuva444p
9669 Swap Y and A plane in yuva444p stream:
9671 format=yuva444p,mergeplanes=0x03010200:yuva444p
9675 Swap U and V plane in yuv420p stream:
9677 format=yuv420p,mergeplanes=0x000201:yuv420p
9681 Cast a rgb24 clip to yuv444p:
9683 format=rgb24,mergeplanes=0x000102:yuv444p
9689 Estimate and export motion vectors using block matching algorithms.
9690 Motion vectors are stored in frame side data to be used by other filters.
9692 This filter accepts the following options:
9695 Specify the motion estimation method. Accepts one of the following values:
9699 Exhaustive search algorithm.
9701 Three step search algorithm.
9703 Two dimensional logarithmic search algorithm.
9705 New three step search algorithm.
9707 Four step search algorithm.
9709 Diamond search algorithm.
9711 Hexagon-based search algorithm.
9713 Enhanced predictive zonal search algorithm.
9715 Uneven multi-hexagon search algorithm.
9717 Default value is @samp{esa}.
9720 Macroblock size. Default @code{16}.
9723 Search parameter. Default @code{7}.
9726 @section midequalizer
9728 Apply Midway Image Equalization effect using two video streams.
9730 Midway Image Equalization adjusts a pair of images to have the same
9731 histogram, while maintaining their dynamics as much as possible. It's
9732 useful for e.g. matching exposures from a pair of stereo cameras.
9734 This filter has two inputs and one output, which must be of same pixel format, but
9735 may be of different sizes. The output of filter is first input adjusted with
9736 midway histogram of both inputs.
9738 This filter accepts the following option:
9742 Set which planes to process. Default is @code{15}, which is all available planes.
9745 @section minterpolate
9747 Convert the video to specified frame rate using motion interpolation.
9749 This filter accepts the following options:
9752 Specify the output frame rate. This can be rational e.g. @code{60000/1001}. Frames are dropped if @var{fps} is lower than source fps. Default @code{60}.
9755 Motion interpolation mode. Following values are accepted:
9758 Duplicate previous or next frame for interpolating new ones.
9760 Blend source frames. Interpolated frame is mean of previous and next frames.
9762 Motion compensated interpolation. Following options are effective when this mode is selected:
9766 Motion compensation mode. Following values are accepted:
9769 Overlapped block motion compensation.
9771 Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9773 Default mode is @samp{obmc}.
9776 Motion estimation mode. Following values are accepted:
9779 Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9781 Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9783 Default mode is @samp{bilat}.
9786 The algorithm to be used for motion estimation. Following values are accepted:
9789 Exhaustive search algorithm.
9791 Three step search algorithm.
9793 Two dimensional logarithmic search algorithm.
9795 New three step search algorithm.
9797 Four step search algorithm.
9799 Diamond search algorithm.
9801 Hexagon-based search algorithm.
9803 Enhanced predictive zonal search algorithm.
9805 Uneven multi-hexagon search algorithm.
9807 Default algorithm is @samp{epzs}.
9810 Macroblock size. Default @code{16}.
9813 Motion estimation search parameter. Default @code{32}.
9816 Enable variable-size block motion compensation. Motion estimation is applied with smaller block sizes at object boundaries in order to make the them less blur. Default is @code{0} (disabled).
9821 Scene change detection method. Scene change leads motion vectors to be in random direction. Scene change detection replace interpolated frames by duplicate ones. May not be needed for other modes. Following values are accepted:
9824 Disable scene change detection.
9826 Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9828 Default method is @samp{fdiff}.
9831 Scene change detection threshold. Default is @code{5.0}.
9836 Drop frames that do not differ greatly from the previous frame in
9837 order to reduce frame rate.
9839 The main use of this filter is for very-low-bitrate encoding
9840 (e.g. streaming over dialup modem), but it could in theory be used for
9841 fixing movies that were inverse-telecined incorrectly.
9843 A description of the accepted options follows.
9847 Set the maximum number of consecutive frames which can be dropped (if
9848 positive), or the minimum interval between dropped frames (if
9849 negative). If the value is 0, the frame is dropped unregarding the
9850 number of previous sequentially dropped frames.
9857 Set the dropping threshold values.
9859 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9860 represent actual pixel value differences, so a threshold of 64
9861 corresponds to 1 unit of difference for each pixel, or the same spread
9862 out differently over the block.
9864 A frame is a candidate for dropping if no 8x8 blocks differ by more
9865 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9866 meaning the whole image) differ by more than a threshold of @option{lo}.
9868 Default value for @option{hi} is 64*12, default value for @option{lo} is
9869 64*5, and default value for @option{frac} is 0.33.
9877 It accepts an integer in input; if non-zero it negates the
9878 alpha component (if available). The default value in input is 0.
9882 Denoise frames using Non-Local Means algorithm.
9884 Each pixel is adjusted by looking for other pixels with similar contexts. This
9885 context similarity is defined by comparing their surrounding patches of size
9886 @option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9889 Note that the research area defines centers for patches, which means some
9890 patches will be made of pixels outside that research area.
9892 The filter accepts the following options.
9896 Set denoising strength.
9902 Same as @option{p} but for chroma planes.
9904 The default value is @var{0} and means automatic.
9910 Same as @option{r} but for chroma planes.
9912 The default value is @var{0} and means automatic.
9917 Deinterlace video using neural network edge directed interpolation.
9919 This filter accepts the following options:
9923 Mandatory option, without binary file filter can not work.
9924 Currently file can be found here:
9925 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9928 Set which frames to deinterlace, by default it is @code{all}.
9929 Can be @code{all} or @code{interlaced}.
9932 Set mode of operation.
9934 Can be one of the following:
9938 Use frame flags, both fields.
9940 Use frame flags, single field.
9944 Use bottom field only.
9946 Use both fields, top first.
9948 Use both fields, bottom first.
9952 Set which planes to process, by default filter process all frames.
9955 Set size of local neighborhood around each pixel, used by the predictor neural
9958 Can be one of the following:
9971 Set the number of neurons in predicctor neural network.
9972 Can be one of the following:
9983 Controls the number of different neural network predictions that are blended
9984 together to compute the final output value. Can be @code{fast}, default or
9988 Set which set of weights to use in the predictor.
9989 Can be one of the following:
9993 weights trained to minimize absolute error
9995 weights trained to minimize squared error
9999 Controls whether or not the prescreener neural network is used to decide
10000 which pixels should be processed by the predictor neural network and which
10001 can be handled by simple cubic interpolation.
10002 The prescreener is trained to know whether cubic interpolation will be
10003 sufficient for a pixel or whether it should be predicted by the predictor nn.
10004 The computational complexity of the prescreener nn is much less than that of
10005 the predictor nn. Since most pixels can be handled by cubic interpolation,
10006 using the prescreener generally results in much faster processing.
10007 The prescreener is pretty accurate, so the difference between using it and not
10008 using it is almost always unnoticeable.
10010 Can be one of the following:
10018 Default is @code{new}.
10021 Set various debugging flags.
10026 Force libavfilter not to use any of the specified pixel formats for the
10027 input to the next filter.
10029 It accepts the following parameters:
10033 A '|'-separated list of pixel format names, such as
10034 apix_fmts=yuv420p|monow|rgb24".
10038 @subsection Examples
10042 Force libavfilter to use a format different from @var{yuv420p} for the
10043 input to the vflip filter:
10045 noformat=pix_fmts=yuv420p,vflip
10049 Convert the input video to any of the formats not contained in the list:
10051 noformat=yuv420p|yuv444p|yuv410p
10057 Add noise on video input frame.
10059 The filter accepts the following options:
10067 Set noise seed for specific pixel component or all pixel components in case
10068 of @var{all_seed}. Default value is @code{123457}.
10070 @item all_strength, alls
10071 @item c0_strength, c0s
10072 @item c1_strength, c1s
10073 @item c2_strength, c2s
10074 @item c3_strength, c3s
10075 Set noise strength for specific pixel component or all pixel components in case
10076 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
10078 @item all_flags, allf
10079 @item c0_flags, c0f
10080 @item c1_flags, c1f
10081 @item c2_flags, c2f
10082 @item c3_flags, c3f
10083 Set pixel component flags or set flags for all components if @var{all_flags}.
10084 Available values for component flags are:
10087 averaged temporal noise (smoother)
10089 mix random noise with a (semi)regular pattern
10091 temporal noise (noise pattern changes between frames)
10093 uniform noise (gaussian otherwise)
10097 @subsection Examples
10099 Add temporal and uniform noise to input video:
10101 noise=alls=20:allf=t+u
10106 Pass the video source unchanged to the output.
10109 Optical Character Recognition
10111 This filter uses Tesseract for optical character recognition.
10113 It accepts the following options:
10117 Set datapath to tesseract data. Default is to use whatever was
10118 set at installation.
10121 Set language, default is "eng".
10124 Set character whitelist.
10127 Set character blacklist.
10130 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10134 Apply a video transform using libopencv.
10136 To enable this filter, install the libopencv library and headers and
10137 configure FFmpeg with @code{--enable-libopencv}.
10139 It accepts the following parameters:
10144 The name of the libopencv filter to apply.
10146 @item filter_params
10147 The parameters to pass to the libopencv filter. If not specified, the default
10148 values are assumed.
10152 Refer to the official libopencv documentation for more precise
10154 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10156 Several libopencv filters are supported; see the following subsections.
10161 Dilate an image by using a specific structuring element.
10162 It corresponds to the libopencv function @code{cvDilate}.
10164 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10166 @var{struct_el} represents a structuring element, and has the syntax:
10167 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10169 @var{cols} and @var{rows} represent the number of columns and rows of
10170 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10171 point, and @var{shape} the shape for the structuring element. @var{shape}
10172 must be "rect", "cross", "ellipse", or "custom".
10174 If the value for @var{shape} is "custom", it must be followed by a
10175 string of the form "=@var{filename}". The file with name
10176 @var{filename} is assumed to represent a binary image, with each
10177 printable character corresponding to a bright pixel. When a custom
10178 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10179 or columns and rows of the read file are assumed instead.
10181 The default value for @var{struct_el} is "3x3+0x0/rect".
10183 @var{nb_iterations} specifies the number of times the transform is
10184 applied to the image, and defaults to 1.
10188 # Use the default values
10191 # Dilate using a structuring element with a 5x5 cross, iterating two times
10192 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10194 # Read the shape from the file diamond.shape, iterating two times.
10195 # The file diamond.shape may contain a pattern of characters like this
10201 # The specified columns and rows are ignored
10202 # but the anchor point coordinates are not
10203 ocv=dilate:0x0+2x2/custom=diamond.shape|2
10208 Erode an image by using a specific structuring element.
10209 It corresponds to the libopencv function @code{cvErode}.
10211 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10212 with the same syntax and semantics as the @ref{dilate} filter.
10216 Smooth the input video.
10218 The filter takes the following parameters:
10219 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10221 @var{type} is the type of smooth filter to apply, and must be one of
10222 the following values: "blur", "blur_no_scale", "median", "gaussian",
10223 or "bilateral". The default value is "gaussian".
10225 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10226 depend on the smooth type. @var{param1} and
10227 @var{param2} accept integer positive values or 0. @var{param3} and
10228 @var{param4} accept floating point values.
10230 The default value for @var{param1} is 3. The default value for the
10231 other parameters is 0.
10233 These parameters correspond to the parameters assigned to the
10234 libopencv function @code{cvSmooth}.
10236 @section oscilloscope
10238 2D Video Oscilloscope.
10240 Useful to measure spatial impulse, step responses, chroma delays, etc.
10242 It accepts the following parameters:
10246 Set scope center x position.
10249 Set scope center y position.
10252 Set scope size, relative to frame diagonal.
10255 Set scope tilt/rotation.
10261 Set trace center x position.
10264 Set trace center y position.
10267 Set trace width, relative to width of frame.
10270 Set trace height, relative to height of frame.
10273 Set which components to trace. By default it traces first three components.
10276 Draw trace grid. By default is enabled.
10279 Draw some statistics. By default is enabled.
10282 Draw scope. By default is enabled.
10285 @subsection Examples
10289 Inspect full first row of video frame.
10291 oscilloscope=x=0.5:y=0:s=1
10295 Inspect full last row of video frame.
10297 oscilloscope=x=0.5:y=1:s=1
10301 Inspect full 5th line of video frame of height 1080.
10303 oscilloscope=x=0.5:y=5/1080:s=1
10307 Inspect full last column of video frame.
10309 oscilloscope=x=1:y=0.5:s=1:t=1
10317 Overlay one video on top of another.
10319 It takes two inputs and has one output. The first input is the "main"
10320 video on which the second input is overlaid.
10322 It accepts the following parameters:
10324 A description of the accepted options follows.
10329 Set the expression for the x and y coordinates of the overlaid video
10330 on the main video. Default value is "0" for both expressions. In case
10331 the expression is invalid, it is set to a huge value (meaning that the
10332 overlay will not be displayed within the output visible area).
10335 The action to take when EOF is encountered on the secondary input; it accepts
10336 one of the following values:
10340 Repeat the last frame (the default).
10344 Pass the main input through.
10348 Set when the expressions for @option{x}, and @option{y} are evaluated.
10350 It accepts the following values:
10353 only evaluate expressions once during the filter initialization or
10354 when a command is processed
10357 evaluate expressions for each incoming frame
10360 Default value is @samp{frame}.
10363 If set to 1, force the output to terminate when the shortest input
10364 terminates. Default value is 0.
10367 Set the format for the output video.
10369 It accepts the following values:
10372 force YUV420 output
10375 force YUV422 output
10378 force YUV444 output
10381 force packed RGB output
10384 force planar RGB output
10387 Default value is @samp{yuv420}.
10389 @item rgb @emph{(deprecated)}
10390 If set to 1, force the filter to accept inputs in the RGB
10391 color space. Default value is 0. This option is deprecated, use
10392 @option{format} instead.
10395 If set to 1, force the filter to draw the last overlay frame over the
10396 main input until the end of the stream. A value of 0 disables this
10397 behavior. Default value is 1.
10400 The @option{x}, and @option{y} expressions can contain the following
10406 The main input width and height.
10410 The overlay input width and height.
10414 The computed values for @var{x} and @var{y}. They are evaluated for
10419 horizontal and vertical chroma subsample values of the output
10420 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10424 the number of input frame, starting from 0
10427 the position in the file of the input frame, NAN if unknown
10430 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10434 Note that the @var{n}, @var{pos}, @var{t} variables are available only
10435 when evaluation is done @emph{per frame}, and will evaluate to NAN
10436 when @option{eval} is set to @samp{init}.
10438 Be aware that frames are taken from each input video in timestamp
10439 order, hence, if their initial timestamps differ, it is a good idea
10440 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10441 have them begin in the same zero timestamp, as the example for
10442 the @var{movie} filter does.
10444 You can chain together more overlays but you should test the
10445 efficiency of such approach.
10447 @subsection Commands
10449 This filter supports the following commands:
10453 Modify the x and y of the overlay input.
10454 The command accepts the same syntax of the corresponding option.
10456 If the specified expression is not valid, it is kept at its current
10460 @subsection Examples
10464 Draw the overlay at 10 pixels from the bottom right corner of the main
10467 overlay=main_w-overlay_w-10:main_h-overlay_h-10
10470 Using named options the example above becomes:
10472 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10476 Insert a transparent PNG logo in the bottom left corner of the input,
10477 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10479 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10483 Insert 2 different transparent PNG logos (second logo on bottom
10484 right corner) using the @command{ffmpeg} tool:
10486 ffmpeg -i input -i logo1 -i logo2 -filter_complex 'overlay=x=10:y=H-h-10,overlay=x=W-w-10:y=H-h-10' output
10490 Add a transparent color layer on top of the main video; @code{WxH}
10491 must specify the size of the main input to the overlay filter:
10493 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10497 Play an original video and a filtered version (here with the deshake
10498 filter) side by side using the @command{ffplay} tool:
10500 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10503 The above command is the same as:
10505 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10509 Make a sliding overlay appearing from the left to the right top part of the
10510 screen starting since time 2:
10512 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10516 Compose output by putting two input videos side to side:
10518 ffmpeg -i left.avi -i right.avi -filter_complex "
10519 nullsrc=size=200x100 [background];
10520 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10521 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10522 [background][left] overlay=shortest=1 [background+left];
10523 [background+left][right] overlay=shortest=1:x=100 [left+right]
10528 Mask 10-20 seconds of a video by applying the delogo filter to a section
10530 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10531 -vf '[in]split[split_main][split_delogo];[split_delogo]trim=start=360:end=371,delogo=0:0:640:480[delogoed];[split_main][delogoed]overlay=eof_action=pass[out]'
10536 Chain several overlays in cascade:
10538 nullsrc=s=200x200 [bg];
10539 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10540 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10541 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10542 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10543 [in3] null, [mid2] overlay=100:100 [out0]
10550 Apply Overcomplete Wavelet denoiser.
10552 The filter accepts the following options:
10558 Larger depth values will denoise lower frequency components more, but
10559 slow down filtering.
10561 Must be an int in the range 8-16, default is @code{8}.
10563 @item luma_strength, ls
10566 Must be a double value in the range 0-1000, default is @code{1.0}.
10568 @item chroma_strength, cs
10569 Set chroma strength.
10571 Must be a double value in the range 0-1000, default is @code{1.0}.
10577 Add paddings to the input image, and place the original input at the
10578 provided @var{x}, @var{y} coordinates.
10580 It accepts the following parameters:
10585 Specify an expression for the size of the output image with the
10586 paddings added. If the value for @var{width} or @var{height} is 0, the
10587 corresponding input size is used for the output.
10589 The @var{width} expression can reference the value set by the
10590 @var{height} expression, and vice versa.
10592 The default value of @var{width} and @var{height} is 0.
10596 Specify the offsets to place the input image at within the padded area,
10597 with respect to the top/left border of the output image.
10599 The @var{x} expression can reference the value set by the @var{y}
10600 expression, and vice versa.
10602 The default value of @var{x} and @var{y} is 0.
10604 If @var{x} or @var{y} evaluate to a negative number, they'll be changed
10605 so the input image is centered on the padded area.
10608 Specify the color of the padded area. For the syntax of this option,
10609 check the "Color" section in the ffmpeg-utils manual.
10611 The default value of @var{color} is "black".
10614 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10616 It accepts the following values:
10620 Only evaluate expressions once during the filter initialization or when
10621 a command is processed.
10624 Evaluate expressions for each incoming frame.
10628 Default value is @samp{init}.
10631 Pad to aspect instead to a resolution.
10635 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10636 options are expressions containing the following constants:
10641 The input video width and height.
10645 These are the same as @var{in_w} and @var{in_h}.
10649 The output width and height (the size of the padded area), as
10650 specified by the @var{width} and @var{height} expressions.
10654 These are the same as @var{out_w} and @var{out_h}.
10658 The x and y offsets as specified by the @var{x} and @var{y}
10659 expressions, or NAN if not yet specified.
10662 same as @var{iw} / @var{ih}
10665 input sample aspect ratio
10668 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10672 The horizontal and vertical chroma subsample values. For example for the
10673 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10676 @subsection Examples
10680 Add paddings with the color "violet" to the input video. The output video
10681 size is 640x480, and the top-left corner of the input video is placed at
10684 pad=640:480:0:40:violet
10687 The example above is equivalent to the following command:
10689 pad=width=640:height=480:x=0:y=40:color=violet
10693 Pad the input to get an output with dimensions increased by 3/2,
10694 and put the input video at the center of the padded area:
10696 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10700 Pad the input to get a squared output with size equal to the maximum
10701 value between the input width and height, and put the input video at
10702 the center of the padded area:
10704 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10708 Pad the input to get a final w/h ratio of 16:9:
10710 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10714 In case of anamorphic video, in order to set the output display aspect
10715 correctly, it is necessary to use @var{sar} in the expression,
10716 according to the relation:
10718 (ih * X / ih) * sar = output_dar
10719 X = output_dar / sar
10722 Thus the previous example needs to be modified to:
10724 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10728 Double the output size and put the input video in the bottom-right
10729 corner of the output padded area:
10731 pad="2*iw:2*ih:ow-iw:oh-ih"
10735 @anchor{palettegen}
10736 @section palettegen
10738 Generate one palette for a whole video stream.
10740 It accepts the following options:
10744 Set the maximum number of colors to quantize in the palette.
10745 Note: the palette will still contain 256 colors; the unused palette entries
10748 @item reserve_transparent
10749 Create a palette of 255 colors maximum and reserve the last one for
10750 transparency. Reserving the transparency color is useful for GIF optimization.
10751 If not set, the maximum of colors in the palette will be 256. You probably want
10752 to disable this option for a standalone image.
10756 Set statistics mode.
10758 It accepts the following values:
10761 Compute full frame histograms.
10763 Compute histograms only for the part that differs from previous frame. This
10764 might be relevant to give more importance to the moving part of your input if
10765 the background is static.
10767 Compute new histogram for each frame.
10770 Default value is @var{full}.
10773 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10774 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10775 color quantization of the palette. This information is also visible at
10776 @var{info} logging level.
10778 @subsection Examples
10782 Generate a representative palette of a given video using @command{ffmpeg}:
10784 ffmpeg -i input.mkv -vf palettegen palette.png
10788 @section paletteuse
10790 Use a palette to downsample an input video stream.
10792 The filter takes two inputs: one video stream and a palette. The palette must
10793 be a 256 pixels image.
10795 It accepts the following options:
10799 Select dithering mode. Available algorithms are:
10802 Ordered 8x8 bayer dithering (deterministic)
10804 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10805 Note: this dithering is sometimes considered "wrong" and is included as a
10807 @item floyd_steinberg
10808 Floyd and Steingberg dithering (error diffusion)
10810 Frankie Sierra dithering v2 (error diffusion)
10812 Frankie Sierra dithering v2 "Lite" (error diffusion)
10815 Default is @var{sierra2_4a}.
10818 When @var{bayer} dithering is selected, this option defines the scale of the
10819 pattern (how much the crosshatch pattern is visible). A low value means more
10820 visible pattern for less banding, and higher value means less visible pattern
10821 at the cost of more banding.
10823 The option must be an integer value in the range [0,5]. Default is @var{2}.
10826 If set, define the zone to process
10830 Only the changing rectangle will be reprocessed. This is similar to GIF
10831 cropping/offsetting compression mechanism. This option can be useful for speed
10832 if only a part of the image is changing, and has use cases such as limiting the
10833 scope of the error diffusal @option{dither} to the rectangle that bounds the
10834 moving scene (it leads to more deterministic output if the scene doesn't change
10835 much, and as a result less moving noise and better GIF compression).
10838 Default is @var{none}.
10841 Take new palette for each output frame.
10844 @subsection Examples
10848 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10849 using @command{ffmpeg}:
10851 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10855 @section perspective
10857 Correct perspective of video not recorded perpendicular to the screen.
10859 A description of the accepted parameters follows.
10870 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10871 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10872 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10873 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10874 then the corners of the source will be sent to the specified coordinates.
10876 The expressions can use the following variables:
10881 the width and height of video frame.
10885 Output frame count.
10888 @item interpolation
10889 Set interpolation for perspective correction.
10891 It accepts the following values:
10897 Default value is @samp{linear}.
10900 Set interpretation of coordinate options.
10902 It accepts the following values:
10906 Send point in the source specified by the given coordinates to
10907 the corners of the destination.
10909 @item 1, destination
10911 Send the corners of the source to the point in the destination specified
10912 by the given coordinates.
10914 Default value is @samp{source}.
10918 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10920 It accepts the following values:
10923 only evaluate expressions once during the filter initialization or
10924 when a command is processed
10927 evaluate expressions for each incoming frame
10930 Default value is @samp{init}.
10935 Delay interlaced video by one field time so that the field order changes.
10937 The intended use is to fix PAL movies that have been captured with the
10938 opposite field order to the film-to-video transfer.
10940 A description of the accepted parameters follows.
10946 It accepts the following values:
10949 Capture field order top-first, transfer bottom-first.
10950 Filter will delay the bottom field.
10953 Capture field order bottom-first, transfer top-first.
10954 Filter will delay the top field.
10957 Capture and transfer with the same field order. This mode only exists
10958 for the documentation of the other options to refer to, but if you
10959 actually select it, the filter will faithfully do nothing.
10962 Capture field order determined automatically by field flags, transfer
10964 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10965 basis using field flags. If no field information is available,
10966 then this works just like @samp{u}.
10969 Capture unknown or varying, transfer opposite.
10970 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10971 analyzing the images and selecting the alternative that produces best
10972 match between the fields.
10975 Capture top-first, transfer unknown or varying.
10976 Filter selects among @samp{t} and @samp{p} using image analysis.
10979 Capture bottom-first, transfer unknown or varying.
10980 Filter selects among @samp{b} and @samp{p} using image analysis.
10983 Capture determined by field flags, transfer unknown or varying.
10984 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10985 image analysis. If no field information is available, then this works just
10986 like @samp{U}. This is the default mode.
10989 Both capture and transfer unknown or varying.
10990 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10994 @section pixdesctest
10996 Pixel format descriptor test filter, mainly useful for internal
10997 testing. The output video should be equal to the input video.
11001 format=monow, pixdesctest
11004 can be used to test the monowhite pixel format descriptor definition.
11008 Display sample values of color channels. Mainly useful for checking color and levels.
11010 The filters accept the following options:
11014 Set scope X position, offset on X axis.
11017 Set scope Y position, offset on Y axis.
11026 Set window opacity. This window also holds statistics about pixel area.
11031 Enable the specified chain of postprocessing subfilters using libpostproc. This
11032 library should be automatically selected with a GPL build (@code{--enable-gpl}).
11033 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
11034 Each subfilter and some options have a short and a long name that can be used
11035 interchangeably, i.e. dr/dering are the same.
11037 The filters accept the following options:
11041 Set postprocessing subfilters string.
11044 All subfilters share common options to determine their scope:
11048 Honor the quality commands for this subfilter.
11051 Do chrominance filtering, too (default).
11054 Do luminance filtering only (no chrominance).
11057 Do chrominance filtering only (no luminance).
11060 These options can be appended after the subfilter name, separated by a '|'.
11062 Available subfilters are:
11065 @item hb/hdeblock[|difference[|flatness]]
11066 Horizontal deblocking filter
11069 Difference factor where higher values mean more deblocking (default: @code{32}).
11071 Flatness threshold where lower values mean more deblocking (default: @code{39}).
11074 @item vb/vdeblock[|difference[|flatness]]
11075 Vertical deblocking filter
11078 Difference factor where higher values mean more deblocking (default: @code{32}).
11080 Flatness threshold where lower values mean more deblocking (default: @code{39}).
11083 @item ha/hadeblock[|difference[|flatness]]
11084 Accurate horizontal deblocking filter
11087 Difference factor where higher values mean more deblocking (default: @code{32}).
11089 Flatness threshold where lower values mean more deblocking (default: @code{39}).
11092 @item va/vadeblock[|difference[|flatness]]
11093 Accurate vertical deblocking filter
11096 Difference factor where higher values mean more deblocking (default: @code{32}).
11098 Flatness threshold where lower values mean more deblocking (default: @code{39}).
11102 The horizontal and vertical deblocking filters share the difference and
11103 flatness values so you cannot set different horizontal and vertical
11107 @item h1/x1hdeblock
11108 Experimental horizontal deblocking filter
11110 @item v1/x1vdeblock
11111 Experimental vertical deblocking filter
11116 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
11119 larger -> stronger filtering
11121 larger -> stronger filtering
11123 larger -> stronger filtering
11126 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
11129 Stretch luminance to @code{0-255}.
11132 @item lb/linblenddeint
11133 Linear blend deinterlacing filter that deinterlaces the given block by
11134 filtering all lines with a @code{(1 2 1)} filter.
11136 @item li/linipoldeint
11137 Linear interpolating deinterlacing filter that deinterlaces the given block by
11138 linearly interpolating every second line.
11140 @item ci/cubicipoldeint
11141 Cubic interpolating deinterlacing filter deinterlaces the given block by
11142 cubically interpolating every second line.
11144 @item md/mediandeint
11145 Median deinterlacing filter that deinterlaces the given block by applying a
11146 median filter to every second line.
11148 @item fd/ffmpegdeint
11149 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
11150 second line with a @code{(-1 4 2 4 -1)} filter.
11153 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
11154 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
11156 @item fq/forceQuant[|quantizer]
11157 Overrides the quantizer table from the input with the constant quantizer you
11165 Default pp filter combination (@code{hb|a,vb|a,dr|a})
11168 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
11171 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
11174 @subsection Examples
11178 Apply horizontal and vertical deblocking, deringing and automatic
11179 brightness/contrast:
11185 Apply default filters without brightness/contrast correction:
11191 Apply default filters and temporal denoiser:
11193 pp=default/tmpnoise|1|2|3
11197 Apply deblocking on luminance only, and switch vertical deblocking on or off
11198 automatically depending on available CPU time:
11205 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11206 similar to spp = 6 with 7 point DCT, where only the center sample is
11209 The filter accepts the following options:
11213 Force a constant quantization parameter. It accepts an integer in range
11214 0 to 63. If not set, the filter will use the QP from the video stream
11218 Set thresholding mode. Available modes are:
11222 Set hard thresholding.
11224 Set soft thresholding (better de-ringing effect, but likely blurrier).
11226 Set medium thresholding (good results, default).
11230 @section premultiply
11231 Apply alpha premultiply effect to input video stream using first plane
11232 of second stream as alpha.
11234 Both streams must have same dimensions and same pixel format.
11236 The filter accepts the following option:
11240 Set which planes will be processed, unprocessed planes will be copied.
11241 By default value 0xf, all planes will be processed.
11245 Apply prewitt operator to input video stream.
11247 The filter accepts the following option:
11251 Set which planes will be processed, unprocessed planes will be copied.
11252 By default value 0xf, all planes will be processed.
11255 Set value which will be multiplied with filtered result.
11258 Set value which will be added to filtered result.
11263 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11264 Ratio) between two input videos.
11266 This filter takes in input two input videos, the first input is
11267 considered the "main" source and is passed unchanged to the
11268 output. The second input is used as a "reference" video for computing
11271 Both video inputs must have the same resolution and pixel format for
11272 this filter to work correctly. Also it assumes that both inputs
11273 have the same number of frames, which are compared one by one.
11275 The obtained average PSNR is printed through the logging system.
11277 The filter stores the accumulated MSE (mean squared error) of each
11278 frame, and at the end of the processing it is averaged across all frames
11279 equally, and the following formula is applied to obtain the PSNR:
11282 PSNR = 10*log10(MAX^2/MSE)
11285 Where MAX is the average of the maximum values of each component of the
11288 The description of the accepted parameters follows.
11291 @item stats_file, f
11292 If specified the filter will use the named file to save the PSNR of
11293 each individual frame. When filename equals "-" the data is sent to
11296 @item stats_version
11297 Specifies which version of the stats file format to use. Details of
11298 each format are written below.
11299 Default value is 1.
11301 @item stats_add_max
11302 Determines whether the max value is output to the stats log.
11303 Default value is 0.
11304 Requires stats_version >= 2. If this is set and stats_version < 2,
11305 the filter will return an error.
11308 The file printed if @var{stats_file} is selected, contains a sequence of
11309 key/value pairs of the form @var{key}:@var{value} for each compared
11312 If a @var{stats_version} greater than 1 is specified, a header line precedes
11313 the list of per-frame-pair stats, with key value pairs following the frame
11314 format with the following parameters:
11317 @item psnr_log_version
11318 The version of the log file format. Will match @var{stats_version}.
11321 A comma separated list of the per-frame-pair parameters included in
11325 A description of each shown per-frame-pair parameter follows:
11329 sequential number of the input frame, starting from 1
11332 Mean Square Error pixel-by-pixel average difference of the compared
11333 frames, averaged over all the image components.
11335 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11336 Mean Square Error pixel-by-pixel average difference of the compared
11337 frames for the component specified by the suffix.
11339 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11340 Peak Signal to Noise ratio of the compared frames for the component
11341 specified by the suffix.
11343 @item max_avg, max_y, max_u, max_v
11344 Maximum allowed value for each channel, and average over all
11350 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11351 [main][ref] psnr="stats_file=stats.log" [out]
11354 On this example the input file being processed is compared with the
11355 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11356 is stored in @file{stats.log}.
11361 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11362 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11365 The pullup filter is designed to take advantage of future context in making
11366 its decisions. This filter is stateless in the sense that it does not lock
11367 onto a pattern to follow, but it instead looks forward to the following
11368 fields in order to identify matches and rebuild progressive frames.
11370 To produce content with an even framerate, insert the fps filter after
11371 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11372 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11374 The filter accepts the following options:
11381 These options set the amount of "junk" to ignore at the left, right, top, and
11382 bottom of the image, respectively. Left and right are in units of 8 pixels,
11383 while top and bottom are in units of 2 lines.
11384 The default is 8 pixels on each side.
11387 Set the strict breaks. Setting this option to 1 will reduce the chances of
11388 filter generating an occasional mismatched frame, but it may also cause an
11389 excessive number of frames to be dropped during high motion sequences.
11390 Conversely, setting it to -1 will make filter match fields more easily.
11391 This may help processing of video where there is slight blurring between
11392 the fields, but may also cause there to be interlaced frames in the output.
11393 Default value is @code{0}.
11396 Set the metric plane to use. It accepts the following values:
11402 Use chroma blue plane.
11405 Use chroma red plane.
11408 This option may be set to use chroma plane instead of the default luma plane
11409 for doing filter's computations. This may improve accuracy on very clean
11410 source material, but more likely will decrease accuracy, especially if there
11411 is chroma noise (rainbow effect) or any grayscale video.
11412 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11413 load and make pullup usable in realtime on slow machines.
11416 For best results (without duplicated frames in the output file) it is
11417 necessary to change the output frame rate. For example, to inverse
11418 telecine NTSC input:
11420 ffmpeg -i input -vf pullup -r 24000/1001 ...
11425 Change video quantization parameters (QP).
11427 The filter accepts the following option:
11431 Set expression for quantization parameter.
11434 The expression is evaluated through the eval API and can contain, among others,
11435 the following constants:
11439 1 if index is not 129, 0 otherwise.
11442 Sequentional index starting from -129 to 128.
11445 @subsection Examples
11449 Some equation like:
11457 Flush video frames from internal cache of frames into a random order.
11458 No frame is discarded.
11459 Inspired by @ref{frei0r} nervous filter.
11463 Set size in number of frames of internal cache, in range from @code{2} to
11464 @code{512}. Default is @code{30}.
11467 Set seed for random number generator, must be an integer included between
11468 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11469 less than @code{0}, the filter will try to use a good random seed on a
11473 @section readeia608
11475 Read closed captioning (EIA-608) information from the top lines of a video frame.
11477 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11478 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11479 with EIA-608 data (starting from 0). A description of each metadata value follows:
11482 @item lavfi.readeia608.X.cc
11483 The two bytes stored as EIA-608 data (printed in hexadecimal).
11485 @item lavfi.readeia608.X.line
11486 The number of the line on which the EIA-608 data was identified and read.
11489 This filter accepts the following options:
11493 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11496 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11499 Set minimal acceptable amplitude change for sync codes detection.
11500 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11503 Set the ratio of width reserved for sync code detection.
11504 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11507 Set the max peaks height difference for sync code detection.
11508 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11511 Set max peaks period difference for sync code detection.
11512 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11515 Set the first two max start code bits differences.
11516 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11519 Set the minimum ratio of bits height compared to 3rd start code bit.
11520 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11523 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11526 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11529 Enable checking the parity bit. In the event of a parity error, the filter will output
11530 @code{0x00} for that character. Default is false.
11533 @subsection Examples
11537 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11539 ffprobe -f lavfi -i movie=captioned_video.mov,readeia608 -show_entries frame=pkt_pts_time:frame_tags=lavfi.readeia608.0.cc,lavfi.readeia608.1.cc -of csv
11545 Read vertical interval timecode (VITC) information from the top lines of a
11548 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11549 timecode value, if a valid timecode has been detected. Further metadata key
11550 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11551 timecode data has been found or not.
11553 This filter accepts the following options:
11557 Set the maximum number of lines to scan for VITC data. If the value is set to
11558 @code{-1} the full video frame is scanned. Default is @code{45}.
11561 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11562 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11565 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11566 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11569 @subsection Examples
11573 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11574 draw @code{--:--:--:--} as a placeholder:
11576 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11582 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11584 Destination pixel at position (X, Y) will be picked from source (x, y) position
11585 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11586 value for pixel will be used for destination pixel.
11588 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11589 will have Xmap/Ymap video stream dimensions.
11590 Xmap and Ymap input video streams are 16bit depth, single channel.
11592 @section removegrain
11594 The removegrain filter is a spatial denoiser for progressive video.
11598 Set mode for the first plane.
11601 Set mode for the second plane.
11604 Set mode for the third plane.
11607 Set mode for the fourth plane.
11610 Range of mode is from 0 to 24. Description of each mode follows:
11614 Leave input plane unchanged. Default.
11617 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11620 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11623 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11626 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11627 This is equivalent to a median filter.
11630 Line-sensitive clipping giving the minimal change.
11633 Line-sensitive clipping, intermediate.
11636 Line-sensitive clipping, intermediate.
11639 Line-sensitive clipping, intermediate.
11642 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11645 Replaces the target pixel with the closest neighbour.
11648 [1 2 1] horizontal and vertical kernel blur.
11654 Bob mode, interpolates top field from the line where the neighbours
11655 pixels are the closest.
11658 Bob mode, interpolates bottom field from the line where the neighbours
11659 pixels are the closest.
11662 Bob mode, interpolates top field. Same as 13 but with a more complicated
11663 interpolation formula.
11666 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11667 interpolation formula.
11670 Clips the pixel with the minimum and maximum of respectively the maximum and
11671 minimum of each pair of opposite neighbour pixels.
11674 Line-sensitive clipping using opposite neighbours whose greatest distance from
11675 the current pixel is minimal.
11678 Replaces the pixel with the average of its 8 neighbours.
11681 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11684 Clips pixels using the averages of opposite neighbour.
11687 Same as mode 21 but simpler and faster.
11690 Small edge and halo removal, but reputed useless.
11696 @section removelogo
11698 Suppress a TV station logo, using an image file to determine which
11699 pixels comprise the logo. It works by filling in the pixels that
11700 comprise the logo with neighboring pixels.
11702 The filter accepts the following options:
11706 Set the filter bitmap file, which can be any image format supported by
11707 libavformat. The width and height of the image file must match those of the
11708 video stream being processed.
11711 Pixels in the provided bitmap image with a value of zero are not
11712 considered part of the logo, non-zero pixels are considered part of
11713 the logo. If you use white (255) for the logo and black (0) for the
11714 rest, you will be safe. For making the filter bitmap, it is
11715 recommended to take a screen capture of a black frame with the logo
11716 visible, and then using a threshold filter followed by the erode
11717 filter once or twice.
11719 If needed, little splotches can be fixed manually. Remember that if
11720 logo pixels are not covered, the filter quality will be much
11721 reduced. Marking too many pixels as part of the logo does not hurt as
11722 much, but it will increase the amount of blurring needed to cover over
11723 the image and will destroy more information than necessary, and extra
11724 pixels will slow things down on a large logo.
11726 @section repeatfields
11728 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11729 fields based on its value.
11733 Reverse a video clip.
11735 Warning: This filter requires memory to buffer the entire clip, so trimming
11738 @subsection Examples
11742 Take the first 5 seconds of a clip, and reverse it.
11750 Rotate video by an arbitrary angle expressed in radians.
11752 The filter accepts the following options:
11754 A description of the optional parameters follows.
11757 Set an expression for the angle by which to rotate the input video
11758 clockwise, expressed as a number of radians. A negative value will
11759 result in a counter-clockwise rotation. By default it is set to "0".
11761 This expression is evaluated for each frame.
11764 Set the output width expression, default value is "iw".
11765 This expression is evaluated just once during configuration.
11768 Set the output height expression, default value is "ih".
11769 This expression is evaluated just once during configuration.
11772 Enable bilinear interpolation if set to 1, a value of 0 disables
11773 it. Default value is 1.
11776 Set the color used to fill the output area not covered by the rotated
11777 image. For the general syntax of this option, check the "Color" section in the
11778 ffmpeg-utils manual. If the special value "none" is selected then no
11779 background is printed (useful for example if the background is never shown).
11781 Default value is "black".
11784 The expressions for the angle and the output size can contain the
11785 following constants and functions:
11789 sequential number of the input frame, starting from 0. It is always NAN
11790 before the first frame is filtered.
11793 time in seconds of the input frame, it is set to 0 when the filter is
11794 configured. It is always NAN before the first frame is filtered.
11798 horizontal and vertical chroma subsample values. For example for the
11799 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11803 the input video width and height
11807 the output width and height, that is the size of the padded area as
11808 specified by the @var{width} and @var{height} expressions
11812 the minimal width/height required for completely containing the input
11813 video rotated by @var{a} radians.
11815 These are only available when computing the @option{out_w} and
11816 @option{out_h} expressions.
11819 @subsection Examples
11823 Rotate the input by PI/6 radians clockwise:
11829 Rotate the input by PI/6 radians counter-clockwise:
11835 Rotate the input by 45 degrees clockwise:
11841 Apply a constant rotation with period T, starting from an angle of PI/3:
11843 rotate=PI/3+2*PI*t/T
11847 Make the input video rotation oscillating with a period of T
11848 seconds and an amplitude of A radians:
11850 rotate=A*sin(2*PI/T*t)
11854 Rotate the video, output size is chosen so that the whole rotating
11855 input video is always completely contained in the output:
11857 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11861 Rotate the video, reduce the output size so that no background is ever
11864 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11868 @subsection Commands
11870 The filter supports the following commands:
11874 Set the angle expression.
11875 The command accepts the same syntax of the corresponding option.
11877 If the specified expression is not valid, it is kept at its current
11883 Apply Shape Adaptive Blur.
11885 The filter accepts the following options:
11888 @item luma_radius, lr
11889 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11890 value is 1.0. A greater value will result in a more blurred image, and
11891 in slower processing.
11893 @item luma_pre_filter_radius, lpfr
11894 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11897 @item luma_strength, ls
11898 Set luma maximum difference between pixels to still be considered, must
11899 be a value in the 0.1-100.0 range, default value is 1.0.
11901 @item chroma_radius, cr
11902 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11903 greater value will result in a more blurred image, and in slower
11906 @item chroma_pre_filter_radius, cpfr
11907 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11909 @item chroma_strength, cs
11910 Set chroma maximum difference between pixels to still be considered,
11911 must be a value in the -0.9-100.0 range.
11914 Each chroma option value, if not explicitly specified, is set to the
11915 corresponding luma option value.
11920 Scale (resize) the input video, using the libswscale library.
11922 The scale filter forces the output display aspect ratio to be the same
11923 of the input, by changing the output sample aspect ratio.
11925 If the input image format is different from the format requested by
11926 the next filter, the scale filter will convert the input to the
11929 @subsection Options
11930 The filter accepts the following options, or any of the options
11931 supported by the libswscale scaler.
11933 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11934 the complete list of scaler options.
11939 Set the output video dimension expression. Default value is the input
11942 If the value is 0, the input width is used for the output.
11944 If one of the values is -1, the scale filter will use a value that
11945 maintains the aspect ratio of the input image, calculated from the
11946 other specified dimension. If both of them are -1, the input size is
11949 If one of the values is -n with n > 1, the scale filter will also use a value
11950 that maintains the aspect ratio of the input image, calculated from the other
11951 specified dimension. After that it will, however, make sure that the calculated
11952 dimension is divisible by n and adjust the value if necessary.
11954 See below for the list of accepted constants for use in the dimension
11958 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11962 Only evaluate expressions once during the filter initialization or when a command is processed.
11965 Evaluate expressions for each incoming frame.
11969 Default value is @samp{init}.
11973 Set the interlacing mode. It accepts the following values:
11977 Force interlaced aware scaling.
11980 Do not apply interlaced scaling.
11983 Select interlaced aware scaling depending on whether the source frames
11984 are flagged as interlaced or not.
11987 Default value is @samp{0}.
11990 Set libswscale scaling flags. See
11991 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11992 complete list of values. If not explicitly specified the filter applies
11996 @item param0, param1
11997 Set libswscale input parameters for scaling algorithms that need them. See
11998 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11999 complete documentation. If not explicitly specified the filter applies
12005 Set the video size. For the syntax of this option, check the
12006 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12008 @item in_color_matrix
12009 @item out_color_matrix
12010 Set in/output YCbCr color space type.
12012 This allows the autodetected value to be overridden as well as allows forcing
12013 a specific value used for the output and encoder.
12015 If not specified, the color space type depends on the pixel format.
12021 Choose automatically.
12024 Format conforming to International Telecommunication Union (ITU)
12025 Recommendation BT.709.
12028 Set color space conforming to the United States Federal Communications
12029 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
12032 Set color space conforming to:
12036 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
12039 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
12042 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
12047 Set color space conforming to SMPTE ST 240:1999.
12052 Set in/output YCbCr sample range.
12054 This allows the autodetected value to be overridden as well as allows forcing
12055 a specific value used for the output and encoder. If not specified, the
12056 range depends on the pixel format. Possible values:
12060 Choose automatically.
12063 Set full range (0-255 in case of 8-bit luma).
12066 Set "MPEG" range (16-235 in case of 8-bit luma).
12069 @item force_original_aspect_ratio
12070 Enable decreasing or increasing output video width or height if necessary to
12071 keep the original aspect ratio. Possible values:
12075 Scale the video as specified and disable this feature.
12078 The output video dimensions will automatically be decreased if needed.
12081 The output video dimensions will automatically be increased if needed.
12085 One useful instance of this option is that when you know a specific device's
12086 maximum allowed resolution, you can use this to limit the output video to
12087 that, while retaining the aspect ratio. For example, device A allows
12088 1280x720 playback, and your video is 1920x800. Using this option (set it to
12089 decrease) and specifying 1280x720 to the command line makes the output
12092 Please note that this is a different thing than specifying -1 for @option{w}
12093 or @option{h}, you still need to specify the output resolution for this option
12098 The values of the @option{w} and @option{h} options are expressions
12099 containing the following constants:
12104 The input width and height
12108 These are the same as @var{in_w} and @var{in_h}.
12112 The output (scaled) width and height
12116 These are the same as @var{out_w} and @var{out_h}
12119 The same as @var{iw} / @var{ih}
12122 input sample aspect ratio
12125 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
12129 horizontal and vertical input chroma subsample values. For example for the
12130 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12134 horizontal and vertical output chroma subsample values. For example for the
12135 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12138 @subsection Examples
12142 Scale the input video to a size of 200x100
12147 This is equivalent to:
12158 Specify a size abbreviation for the output size:
12163 which can also be written as:
12169 Scale the input to 2x:
12171 scale=w=2*iw:h=2*ih
12175 The above is the same as:
12177 scale=2*in_w:2*in_h
12181 Scale the input to 2x with forced interlaced scaling:
12183 scale=2*iw:2*ih:interl=1
12187 Scale the input to half size:
12189 scale=w=iw/2:h=ih/2
12193 Increase the width, and set the height to the same size:
12199 Seek Greek harmony:
12206 Increase the height, and set the width to 3/2 of the height:
12208 scale=w=3/2*oh:h=3/5*ih
12212 Increase the size, making the size a multiple of the chroma
12215 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12219 Increase the width to a maximum of 500 pixels,
12220 keeping the same aspect ratio as the input:
12222 scale=w='min(500\, iw*3/2):h=-1'
12226 @subsection Commands
12228 This filter supports the following commands:
12232 Set the output video dimension expression.
12233 The command accepts the same syntax of the corresponding option.
12235 If the specified expression is not valid, it is kept at its current
12241 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12242 format conversion on CUDA video frames. Setting the output width and height
12243 works in the same way as for the @var{scale} filter.
12245 The following additional options are accepted:
12248 The pixel format of the output CUDA frames. If set to the string "same" (the
12249 default), the input format will be kept. Note that automatic format negotiation
12250 and conversion is not yet supported for hardware frames
12253 The interpolation algorithm used for resizing. One of the following:
12260 @item cubic2p_bspline
12261 2-parameter cubic (B=1, C=0)
12263 @item cubic2p_catmullrom
12264 2-parameter cubic (B=0, C=1/2)
12266 @item cubic2p_b05c03
12267 2-parameter cubic (B=1/2, C=3/10)
12279 Scale (resize) the input video, based on a reference video.
12281 See the scale filter for available options, scale2ref supports the same but
12282 uses the reference video instead of the main input as basis.
12284 @subsection Examples
12288 Scale a subtitle stream to match the main video in size before overlaying
12290 'scale2ref[b][a];[a][b]overlay'
12294 @anchor{selectivecolor}
12295 @section selectivecolor
12297 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12298 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12299 by the "purity" of the color (that is, how saturated it already is).
12301 This filter is similar to the Adobe Photoshop Selective Color tool.
12303 The filter accepts the following options:
12306 @item correction_method
12307 Select color correction method.
12309 Available values are:
12312 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12315 Specified adjustments are relative to the original component value.
12317 Default is @code{absolute}.
12319 Adjustments for red pixels (pixels where the red component is the maximum)
12321 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12323 Adjustments for green pixels (pixels where the green component is the maximum)
12325 Adjustments for cyan pixels (pixels where the red component is the minimum)
12327 Adjustments for blue pixels (pixels where the blue component is the maximum)
12329 Adjustments for magenta pixels (pixels where the green component is the minimum)
12331 Adjustments for white pixels (pixels where all components are greater than 128)
12333 Adjustments for all pixels except pure black and pure white
12335 Adjustments for black pixels (pixels where all components are lesser than 128)
12337 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12340 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12341 4 space separated floating point adjustment values in the [-1,1] range,
12342 respectively to adjust the amount of cyan, magenta, yellow and black for the
12343 pixels of its range.
12345 @subsection Examples
12349 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12350 increase magenta by 27% in blue areas:
12352 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12356 Use a Photoshop selective color preset:
12358 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12362 @anchor{separatefields}
12363 @section separatefields
12365 The @code{separatefields} takes a frame-based video input and splits
12366 each frame into its components fields, producing a new half height clip
12367 with twice the frame rate and twice the frame count.
12369 This filter use field-dominance information in frame to decide which
12370 of each pair of fields to place first in the output.
12371 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12373 @section setdar, setsar
12375 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12378 This is done by changing the specified Sample (aka Pixel) Aspect
12379 Ratio, according to the following equation:
12381 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12384 Keep in mind that the @code{setdar} filter does not modify the pixel
12385 dimensions of the video frame. Also, the display aspect ratio set by
12386 this filter may be changed by later filters in the filterchain,
12387 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12390 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12391 the filter output video.
12393 Note that as a consequence of the application of this filter, the
12394 output display aspect ratio will change according to the equation
12397 Keep in mind that the sample aspect ratio set by the @code{setsar}
12398 filter may be changed by later filters in the filterchain, e.g. if
12399 another "setsar" or a "setdar" filter is applied.
12401 It accepts the following parameters:
12404 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12405 Set the aspect ratio used by the filter.
12407 The parameter can be a floating point number string, an expression, or
12408 a string of the form @var{num}:@var{den}, where @var{num} and
12409 @var{den} are the numerator and denominator of the aspect ratio. If
12410 the parameter is not specified, it is assumed the value "0".
12411 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12415 Set the maximum integer value to use for expressing numerator and
12416 denominator when reducing the expressed aspect ratio to a rational.
12417 Default value is @code{100}.
12421 The parameter @var{sar} is an expression containing
12422 the following constants:
12426 These are approximated values for the mathematical constants e
12427 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12430 The input width and height.
12433 These are the same as @var{w} / @var{h}.
12436 The input sample aspect ratio.
12439 The input display aspect ratio. It is the same as
12440 (@var{w} / @var{h}) * @var{sar}.
12443 Horizontal and vertical chroma subsample values. For example, for the
12444 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12447 @subsection Examples
12452 To change the display aspect ratio to 16:9, specify one of the following:
12459 To change the sample aspect ratio to 10:11, specify:
12465 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12466 1000 in the aspect ratio reduction, use the command:
12468 setdar=ratio=16/9:max=1000
12476 Force field for the output video frame.
12478 The @code{setfield} filter marks the interlace type field for the
12479 output frames. It does not change the input frame, but only sets the
12480 corresponding property, which affects how the frame is treated by
12481 following filters (e.g. @code{fieldorder} or @code{yadif}).
12483 The filter accepts the following options:
12488 Available values are:
12492 Keep the same field property.
12495 Mark the frame as bottom-field-first.
12498 Mark the frame as top-field-first.
12501 Mark the frame as progressive.
12507 Show a line containing various information for each input video frame.
12508 The input video is not modified.
12510 The shown line contains a sequence of key/value pairs of the form
12511 @var{key}:@var{value}.
12513 The following values are shown in the output:
12517 The (sequential) number of the input frame, starting from 0.
12520 The Presentation TimeStamp of the input frame, expressed as a number of
12521 time base units. The time base unit depends on the filter input pad.
12524 The Presentation TimeStamp of the input frame, expressed as a number of
12528 The position of the frame in the input stream, or -1 if this information is
12529 unavailable and/or meaningless (for example in case of synthetic video).
12532 The pixel format name.
12535 The sample aspect ratio of the input frame, expressed in the form
12536 @var{num}/@var{den}.
12539 The size of the input frame. For the syntax of this option, check the
12540 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12543 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12544 for bottom field first).
12547 This is 1 if the frame is a key frame, 0 otherwise.
12550 The picture type of the input frame ("I" for an I-frame, "P" for a
12551 P-frame, "B" for a B-frame, or "?" for an unknown type).
12552 Also refer to the documentation of the @code{AVPictureType} enum and of
12553 the @code{av_get_picture_type_char} function defined in
12554 @file{libavutil/avutil.h}.
12557 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12559 @item plane_checksum
12560 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12561 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12564 @section showpalette
12566 Displays the 256 colors palette of each frame. This filter is only relevant for
12567 @var{pal8} pixel format frames.
12569 It accepts the following option:
12573 Set the size of the box used to represent one palette color entry. Default is
12574 @code{30} (for a @code{30x30} pixel box).
12577 @section shuffleframes
12579 Reorder and/or duplicate and/or drop video frames.
12581 It accepts the following parameters:
12585 Set the destination indexes of input frames.
12586 This is space or '|' separated list of indexes that maps input frames to output
12587 frames. Number of indexes also sets maximal value that each index may have.
12588 '-1' index have special meaning and that is to drop frame.
12591 The first frame has the index 0. The default is to keep the input unchanged.
12593 @subsection Examples
12597 Swap second and third frame of every three frames of the input:
12599 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12603 Swap 10th and 1st frame of every ten frames of the input:
12605 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12609 @section shuffleplanes
12611 Reorder and/or duplicate video planes.
12613 It accepts the following parameters:
12618 The index of the input plane to be used as the first output plane.
12621 The index of the input plane to be used as the second output plane.
12624 The index of the input plane to be used as the third output plane.
12627 The index of the input plane to be used as the fourth output plane.
12631 The first plane has the index 0. The default is to keep the input unchanged.
12633 @subsection Examples
12637 Swap the second and third planes of the input:
12639 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12643 @anchor{signalstats}
12644 @section signalstats
12645 Evaluate various visual metrics that assist in determining issues associated
12646 with the digitization of analog video media.
12648 By default the filter will log these metadata values:
12652 Display the minimal Y value contained within the input frame. Expressed in
12656 Display the Y value at the 10% percentile within the input frame. Expressed in
12660 Display the average Y value within the input frame. Expressed in range of
12664 Display the Y value at the 90% percentile within the input frame. Expressed in
12668 Display the maximum Y value contained within the input frame. Expressed in
12672 Display the minimal U value contained within the input frame. Expressed in
12676 Display the U value at the 10% percentile within the input frame. Expressed in
12680 Display the average U value within the input frame. Expressed in range of
12684 Display the U value at the 90% percentile within the input frame. Expressed in
12688 Display the maximum U value contained within the input frame. Expressed in
12692 Display the minimal V value contained within the input frame. Expressed in
12696 Display the V value at the 10% percentile within the input frame. Expressed in
12700 Display the average V value within the input frame. Expressed in range of
12704 Display the V value at the 90% percentile within the input frame. Expressed in
12708 Display the maximum V value contained within the input frame. Expressed in
12712 Display the minimal saturation value contained within the input frame.
12713 Expressed in range of [0-~181.02].
12716 Display the saturation value at the 10% percentile within the input frame.
12717 Expressed in range of [0-~181.02].
12720 Display the average saturation value within the input frame. Expressed in range
12724 Display the saturation value at the 90% percentile within the input frame.
12725 Expressed in range of [0-~181.02].
12728 Display the maximum saturation value contained within the input frame.
12729 Expressed in range of [0-~181.02].
12732 Display the median value for hue within the input frame. Expressed in range of
12736 Display the average value for hue within the input frame. Expressed in range of
12740 Display the average of sample value difference between all values of the Y
12741 plane in the current frame and corresponding values of the previous input frame.
12742 Expressed in range of [0-255].
12745 Display the average of sample value difference between all values of the U
12746 plane in the current frame and corresponding values of the previous input frame.
12747 Expressed in range of [0-255].
12750 Display the average of sample value difference between all values of the V
12751 plane in the current frame and corresponding values of the previous input frame.
12752 Expressed in range of [0-255].
12755 Display bit depth of Y plane in current frame.
12756 Expressed in range of [0-16].
12759 Display bit depth of U plane in current frame.
12760 Expressed in range of [0-16].
12763 Display bit depth of V plane in current frame.
12764 Expressed in range of [0-16].
12767 The filter accepts the following options:
12773 @option{stat} specify an additional form of image analysis.
12774 @option{out} output video with the specified type of pixel highlighted.
12776 Both options accept the following values:
12780 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12781 unlike the neighboring pixels of the same field. Examples of temporal outliers
12782 include the results of video dropouts, head clogs, or tape tracking issues.
12785 Identify @var{vertical line repetition}. Vertical line repetition includes
12786 similar rows of pixels within a frame. In born-digital video vertical line
12787 repetition is common, but this pattern is uncommon in video digitized from an
12788 analog source. When it occurs in video that results from the digitization of an
12789 analog source it can indicate concealment from a dropout compensator.
12792 Identify pixels that fall outside of legal broadcast range.
12796 Set the highlight color for the @option{out} option. The default color is
12800 @subsection Examples
12804 Output data of various video metrics:
12806 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12810 Output specific data about the minimum and maximum values of the Y plane per frame:
12812 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12816 Playback video while highlighting pixels that are outside of broadcast range in red.
12818 ffplay example.mov -vf signalstats="out=brng:color=red"
12822 Playback video with signalstats metadata drawn over the frame.
12824 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12827 The contents of signalstat_drawtext.txt used in the command are:
12830 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12831 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12832 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12833 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12841 Calculates the MPEG-7 Video Signature. The filter can handle more than one
12842 input. In this case the matching between the inputs can be calculated additionally.
12843 The filter always passes through the first input. The signature of each stream can
12844 be written into a file.
12846 It accepts the following options:
12850 Enable or disable the matching process.
12852 Available values are:
12856 Disable the calculation of a matching (default).
12858 Calculate the matching for the whole video and output whether the whole video
12859 matches or only parts.
12861 Calculate only until a matching is found or the video ends. Should be faster in
12866 Set the number of inputs. The option value must be a non negative integer.
12867 Default value is 1.
12870 Set the path to which the output is written. If there is more than one input,
12871 the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12872 integer), that will be replaced with the input number. If no filename is
12873 specified, no output will be written. This is the default.
12876 Choose the output format.
12878 Available values are:
12882 Use the specified binary representation (default).
12884 Use the specified xml representation.
12888 Set threshold to detect one word as similar. The option value must be an integer
12889 greater than zero. The default value is 9000.
12892 Set threshold to detect all words as similar. The option value must be an integer
12893 greater than zero. The default value is 60000.
12896 Set threshold to detect frames as similar. The option value must be an integer
12897 greater than zero. The default value is 116.
12900 Set the minimum length of a sequence in frames to recognize it as matching
12901 sequence. The option value must be a non negative integer value.
12902 The default value is 0.
12905 Set the minimum relation, that matching frames to all frames must have.
12906 The option value must be a double value between 0 and 1. The default value is 0.5.
12909 @subsection Examples
12913 To calculate the signature of an input video and store it in signature.bin:
12915 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12919 To detect whether two videos match and store the signatures in XML format in
12920 signature0.xml and signature1.xml:
12922 ffmpeg -i input1.mkv -i input2.mkv -filter_complex "[0:v][1:v] signature=nb_inputs=2:detectmode=full:format=xml:filename=signature%d.xml" -map :v -f null -
12930 Blur the input video without impacting the outlines.
12932 It accepts the following options:
12935 @item luma_radius, lr
12936 Set the luma radius. The option value must be a float number in
12937 the range [0.1,5.0] that specifies the variance of the gaussian filter
12938 used to blur the image (slower if larger). Default value is 1.0.
12940 @item luma_strength, ls
12941 Set the luma strength. The option value must be a float number
12942 in the range [-1.0,1.0] that configures the blurring. A value included
12943 in [0.0,1.0] will blur the image whereas a value included in
12944 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12946 @item luma_threshold, lt
12947 Set the luma threshold used as a coefficient to determine
12948 whether a pixel should be blurred or not. The option value must be an
12949 integer in the range [-30,30]. A value of 0 will filter all the image,
12950 a value included in [0,30] will filter flat areas and a value included
12951 in [-30,0] will filter edges. Default value is 0.
12953 @item chroma_radius, cr
12954 Set the chroma radius. The option value must be a float number in
12955 the range [0.1,5.0] that specifies the variance of the gaussian filter
12956 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12958 @item chroma_strength, cs
12959 Set the chroma strength. The option value must be a float number
12960 in the range [-1.0,1.0] that configures the blurring. A value included
12961 in [0.0,1.0] will blur the image whereas a value included in
12962 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12964 @item chroma_threshold, ct
12965 Set the chroma threshold used as a coefficient to determine
12966 whether a pixel should be blurred or not. The option value must be an
12967 integer in the range [-30,30]. A value of 0 will filter all the image,
12968 a value included in [0,30] will filter flat areas and a value included
12969 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12972 If a chroma option is not explicitly set, the corresponding luma value
12977 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12979 This filter takes in input two input videos, the first input is
12980 considered the "main" source and is passed unchanged to the
12981 output. The second input is used as a "reference" video for computing
12984 Both video inputs must have the same resolution and pixel format for
12985 this filter to work correctly. Also it assumes that both inputs
12986 have the same number of frames, which are compared one by one.
12988 The filter stores the calculated SSIM of each frame.
12990 The description of the accepted parameters follows.
12993 @item stats_file, f
12994 If specified the filter will use the named file to save the SSIM of
12995 each individual frame. When filename equals "-" the data is sent to
12999 The file printed if @var{stats_file} is selected, contains a sequence of
13000 key/value pairs of the form @var{key}:@var{value} for each compared
13003 A description of each shown parameter follows:
13007 sequential number of the input frame, starting from 1
13009 @item Y, U, V, R, G, B
13010 SSIM of the compared frames for the component specified by the suffix.
13013 SSIM of the compared frames for the whole frame.
13016 Same as above but in dB representation.
13021 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
13022 [main][ref] ssim="stats_file=stats.log" [out]
13025 On this example the input file being processed is compared with the
13026 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
13027 is stored in @file{stats.log}.
13029 Another example with both psnr and ssim at same time:
13031 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
13036 Convert between different stereoscopic image formats.
13038 The filters accept the following options:
13042 Set stereoscopic image format of input.
13044 Available values for input image formats are:
13047 side by side parallel (left eye left, right eye right)
13050 side by side crosseye (right eye left, left eye right)
13053 side by side parallel with half width resolution
13054 (left eye left, right eye right)
13057 side by side crosseye with half width resolution
13058 (right eye left, left eye right)
13061 above-below (left eye above, right eye below)
13064 above-below (right eye above, left eye below)
13067 above-below with half height resolution
13068 (left eye above, right eye below)
13071 above-below with half height resolution
13072 (right eye above, left eye below)
13075 alternating frames (left eye first, right eye second)
13078 alternating frames (right eye first, left eye second)
13081 interleaved rows (left eye has top row, right eye starts on next row)
13084 interleaved rows (right eye has top row, left eye starts on next row)
13087 interleaved columns, left eye first
13090 interleaved columns, right eye first
13092 Default value is @samp{sbsl}.
13096 Set stereoscopic image format of output.
13100 side by side parallel (left eye left, right eye right)
13103 side by side crosseye (right eye left, left eye right)
13106 side by side parallel with half width resolution
13107 (left eye left, right eye right)
13110 side by side crosseye with half width resolution
13111 (right eye left, left eye right)
13114 above-below (left eye above, right eye below)
13117 above-below (right eye above, left eye below)
13120 above-below with half height resolution
13121 (left eye above, right eye below)
13124 above-below with half height resolution
13125 (right eye above, left eye below)
13128 alternating frames (left eye first, right eye second)
13131 alternating frames (right eye first, left eye second)
13134 interleaved rows (left eye has top row, right eye starts on next row)
13137 interleaved rows (right eye has top row, left eye starts on next row)
13140 anaglyph red/blue gray
13141 (red filter on left eye, blue filter on right eye)
13144 anaglyph red/green gray
13145 (red filter on left eye, green filter on right eye)
13148 anaglyph red/cyan gray
13149 (red filter on left eye, cyan filter on right eye)
13152 anaglyph red/cyan half colored
13153 (red filter on left eye, cyan filter on right eye)
13156 anaglyph red/cyan color
13157 (red filter on left eye, cyan filter on right eye)
13160 anaglyph red/cyan color optimized with the least squares projection of dubois
13161 (red filter on left eye, cyan filter on right eye)
13164 anaglyph green/magenta gray
13165 (green filter on left eye, magenta filter on right eye)
13168 anaglyph green/magenta half colored
13169 (green filter on left eye, magenta filter on right eye)
13172 anaglyph green/magenta colored
13173 (green filter on left eye, magenta filter on right eye)
13176 anaglyph green/magenta color optimized with the least squares projection of dubois
13177 (green filter on left eye, magenta filter on right eye)
13180 anaglyph yellow/blue gray
13181 (yellow filter on left eye, blue filter on right eye)
13184 anaglyph yellow/blue half colored
13185 (yellow filter on left eye, blue filter on right eye)
13188 anaglyph yellow/blue colored
13189 (yellow filter on left eye, blue filter on right eye)
13192 anaglyph yellow/blue color optimized with the least squares projection of dubois
13193 (yellow filter on left eye, blue filter on right eye)
13196 mono output (left eye only)
13199 mono output (right eye only)
13202 checkerboard, left eye first
13205 checkerboard, right eye first
13208 interleaved columns, left eye first
13211 interleaved columns, right eye first
13217 Default value is @samp{arcd}.
13220 @subsection Examples
13224 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13230 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13236 @section streamselect, astreamselect
13237 Select video or audio streams.
13239 The filter accepts the following options:
13243 Set number of inputs. Default is 2.
13246 Set input indexes to remap to outputs.
13249 @subsection Commands
13251 The @code{streamselect} and @code{astreamselect} filter supports the following
13256 Set input indexes to remap to outputs.
13259 @subsection Examples
13263 Select first 5 seconds 1st stream and rest of time 2nd stream:
13265 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13269 Same as above, but for audio:
13271 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13276 Apply sobel operator to input video stream.
13278 The filter accepts the following option:
13282 Set which planes will be processed, unprocessed planes will be copied.
13283 By default value 0xf, all planes will be processed.
13286 Set value which will be multiplied with filtered result.
13289 Set value which will be added to filtered result.
13295 Apply a simple postprocessing filter that compresses and decompresses the image
13296 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13297 and average the results.
13299 The filter accepts the following options:
13303 Set quality. This option defines the number of levels for averaging. It accepts
13304 an integer in the range 0-6. If set to @code{0}, the filter will have no
13305 effect. A value of @code{6} means the higher quality. For each increment of
13306 that value the speed drops by a factor of approximately 2. Default value is
13310 Force a constant quantization parameter. If not set, the filter will use the QP
13311 from the video stream (if available).
13314 Set thresholding mode. Available modes are:
13318 Set hard thresholding (default).
13320 Set soft thresholding (better de-ringing effect, but likely blurrier).
13323 @item use_bframe_qp
13324 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13325 option may cause flicker since the B-Frames have often larger QP. Default is
13326 @code{0} (not enabled).
13332 Draw subtitles on top of input video using the libass library.
13334 To enable compilation of this filter you need to configure FFmpeg with
13335 @code{--enable-libass}. This filter also requires a build with libavcodec and
13336 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13337 Alpha) subtitles format.
13339 The filter accepts the following options:
13343 Set the filename of the subtitle file to read. It must be specified.
13345 @item original_size
13346 Specify the size of the original video, the video for which the ASS file
13347 was composed. For the syntax of this option, check the
13348 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13349 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13350 correctly scale the fonts if the aspect ratio has been changed.
13353 Set a directory path containing fonts that can be used by the filter.
13354 These fonts will be used in addition to whatever the font provider uses.
13357 Set subtitles input character encoding. @code{subtitles} filter only. Only
13358 useful if not UTF-8.
13360 @item stream_index, si
13361 Set subtitles stream index. @code{subtitles} filter only.
13364 Override default style or script info parameters of the subtitles. It accepts a
13365 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13368 If the first key is not specified, it is assumed that the first value
13369 specifies the @option{filename}.
13371 For example, to render the file @file{sub.srt} on top of the input
13372 video, use the command:
13377 which is equivalent to:
13379 subtitles=filename=sub.srt
13382 To render the default subtitles stream from file @file{video.mkv}, use:
13384 subtitles=video.mkv
13387 To render the second subtitles stream from that file, use:
13389 subtitles=video.mkv:si=1
13392 To make the subtitles stream from @file{sub.srt} appear in transparent green
13393 @code{DejaVu Serif}, use:
13395 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13398 @section super2xsai
13400 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13401 Interpolate) pixel art scaling algorithm.
13403 Useful for enlarging pixel art images without reducing sharpness.
13407 Swap two rectangular objects in video.
13409 This filter accepts the following options:
13419 Set 1st rect x coordinate.
13422 Set 1st rect y coordinate.
13425 Set 2nd rect x coordinate.
13428 Set 2nd rect y coordinate.
13430 All expressions are evaluated once for each frame.
13433 The all options are expressions containing the following constants:
13438 The input width and height.
13441 same as @var{w} / @var{h}
13444 input sample aspect ratio
13447 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13450 The number of the input frame, starting from 0.
13453 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13456 the position in the file of the input frame, NAN if unknown
13464 Apply telecine process to the video.
13466 This filter accepts the following options:
13475 The default value is @code{top}.
13479 A string of numbers representing the pulldown pattern you wish to apply.
13480 The default value is @code{23}.
13484 Some typical patterns:
13489 24p: 2332 (preferred)
13496 24p: 222222222223 ("Euro pulldown")
13503 Apply threshold effect to video stream.
13505 This filter needs four video streams to perform thresholding.
13506 First stream is stream we are filtering.
13507 Second stream is holding threshold values, third stream is holding min values,
13508 and last, fourth stream is holding max values.
13510 The filter accepts the following option:
13514 Set which planes will be processed, unprocessed planes will be copied.
13515 By default value 0xf, all planes will be processed.
13518 For example if first stream pixel's component value is less then threshold value
13519 of pixel component from 2nd threshold stream, third stream value will picked,
13520 otherwise fourth stream pixel component value will be picked.
13522 Using color source filter one can perform various types of thresholding:
13524 @subsection Examples
13528 Binary threshold, using gray color as threshold:
13530 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13534 Inverted binary threshold, using gray color as threshold:
13536 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13540 Truncate binary threshold, using gray color as threshold:
13542 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13546 Threshold to zero, using gray color as threshold:
13548 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13552 Inverted threshold to zero, using gray color as threshold:
13554 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13559 Select the most representative frame in a given sequence of consecutive frames.
13561 The filter accepts the following options:
13565 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13566 will pick one of them, and then handle the next batch of @var{n} frames until
13567 the end. Default is @code{100}.
13570 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13571 value will result in a higher memory usage, so a high value is not recommended.
13573 @subsection Examples
13577 Extract one picture each 50 frames:
13583 Complete example of a thumbnail creation with @command{ffmpeg}:
13585 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13591 Tile several successive frames together.
13593 The filter accepts the following options:
13598 Set the grid size (i.e. the number of lines and columns). For the syntax of
13599 this option, check the
13600 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13603 Set the maximum number of frames to render in the given area. It must be less
13604 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13605 the area will be used.
13608 Set the outer border margin in pixels.
13611 Set the inner border thickness (i.e. the number of pixels between frames). For
13612 more advanced padding options (such as having different values for the edges),
13613 refer to the pad video filter.
13616 Specify the color of the unused area. For the syntax of this option, check the
13617 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13621 @subsection Examples
13625 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13627 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13629 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13630 duplicating each output frame to accommodate the originally detected frame
13634 Display @code{5} pictures in an area of @code{3x2} frames,
13635 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13636 mixed flat and named options:
13638 tile=3x2:nb_frames=5:padding=7:margin=2
13642 @section tinterlace
13644 Perform various types of temporal field interlacing.
13646 Frames are counted starting from 1, so the first input frame is
13649 The filter accepts the following options:
13654 Specify the mode of the interlacing. This option can also be specified
13655 as a value alone. See below for a list of values for this option.
13657 Available values are:
13661 Move odd frames into the upper field, even into the lower field,
13662 generating a double height frame at half frame rate.
13666 Frame 1 Frame 2 Frame 3 Frame 4
13668 11111 22222 33333 44444
13669 11111 22222 33333 44444
13670 11111 22222 33333 44444
13671 11111 22222 33333 44444
13685 Only output odd frames, even frames are dropped, generating a frame with
13686 unchanged height at half frame rate.
13691 Frame 1 Frame 2 Frame 3 Frame 4
13693 11111 22222 33333 44444
13694 11111 22222 33333 44444
13695 11111 22222 33333 44444
13696 11111 22222 33333 44444
13706 Only output even frames, odd frames are dropped, generating a frame with
13707 unchanged height at half frame rate.
13712 Frame 1 Frame 2 Frame 3 Frame 4
13714 11111 22222 33333 44444
13715 11111 22222 33333 44444
13716 11111 22222 33333 44444
13717 11111 22222 33333 44444
13727 Expand each frame to full height, but pad alternate lines with black,
13728 generating a frame with double height at the same input frame rate.
13733 Frame 1 Frame 2 Frame 3 Frame 4
13735 11111 22222 33333 44444
13736 11111 22222 33333 44444
13737 11111 22222 33333 44444
13738 11111 22222 33333 44444
13741 11111 ..... 33333 .....
13742 ..... 22222 ..... 44444
13743 11111 ..... 33333 .....
13744 ..... 22222 ..... 44444
13745 11111 ..... 33333 .....
13746 ..... 22222 ..... 44444
13747 11111 ..... 33333 .....
13748 ..... 22222 ..... 44444
13752 @item interleave_top, 4
13753 Interleave the upper field from odd frames with the lower field from
13754 even frames, generating a frame with unchanged height at half frame rate.
13759 Frame 1 Frame 2 Frame 3 Frame 4
13761 11111<- 22222 33333<- 44444
13762 11111 22222<- 33333 44444<-
13763 11111<- 22222 33333<- 44444
13764 11111 22222<- 33333 44444<-
13774 @item interleave_bottom, 5
13775 Interleave the lower field from odd frames with the upper field from
13776 even frames, generating a frame with unchanged height at half frame rate.
13781 Frame 1 Frame 2 Frame 3 Frame 4
13783 11111 22222<- 33333 44444<-
13784 11111<- 22222 33333<- 44444
13785 11111 22222<- 33333 44444<-
13786 11111<- 22222 33333<- 44444
13796 @item interlacex2, 6
13797 Double frame rate with unchanged height. Frames are inserted each
13798 containing the second temporal field from the previous input frame and
13799 the first temporal field from the next input frame. This mode relies on
13800 the top_field_first flag. Useful for interlaced video displays with no
13801 field synchronisation.
13806 Frame 1 Frame 2 Frame 3 Frame 4
13808 11111 22222 33333 44444
13809 11111 22222 33333 44444
13810 11111 22222 33333 44444
13811 11111 22222 33333 44444
13814 11111 22222 22222 33333 33333 44444 44444
13815 11111 11111 22222 22222 33333 33333 44444
13816 11111 22222 22222 33333 33333 44444 44444
13817 11111 11111 22222 22222 33333 33333 44444
13822 Move odd frames into the upper field, even into the lower field,
13823 generating a double height frame at same frame rate.
13828 Frame 1 Frame 2 Frame 3 Frame 4
13830 11111 22222 33333 44444
13831 11111 22222 33333 44444
13832 11111 22222 33333 44444
13833 11111 22222 33333 44444
13836 11111 33333 33333 55555
13837 22222 22222 44444 44444
13838 11111 33333 33333 55555
13839 22222 22222 44444 44444
13840 11111 33333 33333 55555
13841 22222 22222 44444 44444
13842 11111 33333 33333 55555
13843 22222 22222 44444 44444
13848 Numeric values are deprecated but are accepted for backward
13849 compatibility reasons.
13851 Default mode is @code{merge}.
13854 Specify flags influencing the filter process.
13856 Available value for @var{flags} is:
13859 @item low_pass_filter, vlfp
13860 Enable vertical low-pass filtering in the filter.
13861 Vertical low-pass filtering is required when creating an interlaced
13862 destination from a progressive source which contains high-frequency
13863 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13866 Vertical low-pass filtering can only be enabled for @option{mode}
13867 @var{interleave_top} and @var{interleave_bottom}.
13874 Transpose rows with columns in the input video and optionally flip it.
13876 It accepts the following parameters:
13881 Specify the transposition direction.
13883 Can assume the following values:
13885 @item 0, 4, cclock_flip
13886 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13894 Rotate by 90 degrees clockwise, that is:
13902 Rotate by 90 degrees counterclockwise, that is:
13909 @item 3, 7, clock_flip
13910 Rotate by 90 degrees clockwise and vertically flip, that is:
13918 For values between 4-7, the transposition is only done if the input
13919 video geometry is portrait and not landscape. These values are
13920 deprecated, the @code{passthrough} option should be used instead.
13922 Numerical values are deprecated, and should be dropped in favor of
13923 symbolic constants.
13926 Do not apply the transposition if the input geometry matches the one
13927 specified by the specified value. It accepts the following values:
13930 Always apply transposition.
13932 Preserve portrait geometry (when @var{height} >= @var{width}).
13934 Preserve landscape geometry (when @var{width} >= @var{height}).
13937 Default value is @code{none}.
13940 For example to rotate by 90 degrees clockwise and preserve portrait
13943 transpose=dir=1:passthrough=portrait
13946 The command above can also be specified as:
13948 transpose=1:portrait
13952 Trim the input so that the output contains one continuous subpart of the input.
13954 It accepts the following parameters:
13957 Specify the time of the start of the kept section, i.e. the frame with the
13958 timestamp @var{start} will be the first frame in the output.
13961 Specify the time of the first frame that will be dropped, i.e. the frame
13962 immediately preceding the one with the timestamp @var{end} will be the last
13963 frame in the output.
13966 This is the same as @var{start}, except this option sets the start timestamp
13967 in timebase units instead of seconds.
13970 This is the same as @var{end}, except this option sets the end timestamp
13971 in timebase units instead of seconds.
13974 The maximum duration of the output in seconds.
13977 The number of the first frame that should be passed to the output.
13980 The number of the first frame that should be dropped.
13983 @option{start}, @option{end}, and @option{duration} are expressed as time
13984 duration specifications; see
13985 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13986 for the accepted syntax.
13988 Note that the first two sets of the start/end options and the @option{duration}
13989 option look at the frame timestamp, while the _frame variants simply count the
13990 frames that pass through the filter. Also note that this filter does not modify
13991 the timestamps. If you wish for the output timestamps to start at zero, insert a
13992 setpts filter after the trim filter.
13994 If multiple start or end options are set, this filter tries to be greedy and
13995 keep all the frames that match at least one of the specified constraints. To keep
13996 only the part that matches all the constraints at once, chain multiple trim
13999 The defaults are such that all the input is kept. So it is possible to set e.g.
14000 just the end values to keep everything before the specified time.
14005 Drop everything except the second minute of input:
14007 ffmpeg -i INPUT -vf trim=60:120
14011 Keep only the first second:
14013 ffmpeg -i INPUT -vf trim=duration=1
14022 Sharpen or blur the input video.
14024 It accepts the following parameters:
14027 @item luma_msize_x, lx
14028 Set the luma matrix horizontal size. It must be an odd integer between
14029 3 and 23. The default value is 5.
14031 @item luma_msize_y, ly
14032 Set the luma matrix vertical size. It must be an odd integer between 3
14033 and 23. The default value is 5.
14035 @item luma_amount, la
14036 Set the luma effect strength. It must be a floating point number, reasonable
14037 values lay between -1.5 and 1.5.
14039 Negative values will blur the input video, while positive values will
14040 sharpen it, a value of zero will disable the effect.
14042 Default value is 1.0.
14044 @item chroma_msize_x, cx
14045 Set the chroma matrix horizontal size. It must be an odd integer
14046 between 3 and 23. The default value is 5.
14048 @item chroma_msize_y, cy
14049 Set the chroma matrix vertical size. It must be an odd integer
14050 between 3 and 23. The default value is 5.
14052 @item chroma_amount, ca
14053 Set the chroma effect strength. It must be a floating point number, reasonable
14054 values lay between -1.5 and 1.5.
14056 Negative values will blur the input video, while positive values will
14057 sharpen it, a value of zero will disable the effect.
14059 Default value is 0.0.
14062 If set to 1, specify using OpenCL capabilities, only available if
14063 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
14067 All parameters are optional and default to the equivalent of the
14068 string '5:5:1.0:5:5:0.0'.
14070 @subsection Examples
14074 Apply strong luma sharpen effect:
14076 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
14080 Apply a strong blur of both luma and chroma parameters:
14082 unsharp=7:7:-2:7:7:-2
14088 Apply ultra slow/simple postprocessing filter that compresses and decompresses
14089 the image at several (or - in the case of @option{quality} level @code{8} - all)
14090 shifts and average the results.
14092 The way this differs from the behavior of spp is that uspp actually encodes &
14093 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
14094 DCT similar to MJPEG.
14096 The filter accepts the following options:
14100 Set quality. This option defines the number of levels for averaging. It accepts
14101 an integer in the range 0-8. If set to @code{0}, the filter will have no
14102 effect. A value of @code{8} means the higher quality. For each increment of
14103 that value the speed drops by a factor of approximately 2. Default value is
14107 Force a constant quantization parameter. If not set, the filter will use the QP
14108 from the video stream (if available).
14111 @section vaguedenoiser
14113 Apply a wavelet based denoiser.
14115 It transforms each frame from the video input into the wavelet domain,
14116 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
14117 the obtained coefficients. It does an inverse wavelet transform after.
14118 Due to wavelet properties, it should give a nice smoothed result, and
14119 reduced noise, without blurring picture features.
14121 This filter accepts the following options:
14125 The filtering strength. The higher, the more filtered the video will be.
14126 Hard thresholding can use a higher threshold than soft thresholding
14127 before the video looks overfiltered.
14130 The filtering method the filter will use.
14132 It accepts the following values:
14135 All values under the threshold will be zeroed.
14138 All values under the threshold will be zeroed. All values above will be
14139 reduced by the threshold.
14142 Scales or nullifies coefficients - intermediary between (more) soft and
14143 (less) hard thresholding.
14147 Number of times, the wavelet will decompose the picture. Picture can't
14148 be decomposed beyond a particular point (typically, 8 for a 640x480
14149 frame - as 2^9 = 512 > 480)
14152 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
14155 A list of the planes to process. By default all planes are processed.
14158 @section vectorscope
14160 Display 2 color component values in the two dimensional graph (which is called
14163 This filter accepts the following options:
14167 Set vectorscope mode.
14169 It accepts the following values:
14172 Gray values are displayed on graph, higher brightness means more pixels have
14173 same component color value on location in graph. This is the default mode.
14176 Gray values are displayed on graph. Surrounding pixels values which are not
14177 present in video frame are drawn in gradient of 2 color components which are
14178 set by option @code{x} and @code{y}. The 3rd color component is static.
14181 Actual color components values present in video frame are displayed on graph.
14184 Similar as color2 but higher frequency of same values @code{x} and @code{y}
14185 on graph increases value of another color component, which is luminance by
14186 default values of @code{x} and @code{y}.
14189 Actual colors present in video frame are displayed on graph. If two different
14190 colors map to same position on graph then color with higher value of component
14191 not present in graph is picked.
14194 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
14195 component picked from radial gradient.
14199 Set which color component will be represented on X-axis. Default is @code{1}.
14202 Set which color component will be represented on Y-axis. Default is @code{2}.
14205 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
14206 of color component which represents frequency of (X, Y) location in graph.
14211 No envelope, this is default.
14214 Instant envelope, even darkest single pixel will be clearly highlighted.
14217 Hold maximum and minimum values presented in graph over time. This way you
14218 can still spot out of range values without constantly looking at vectorscope.
14221 Peak and instant envelope combined together.
14225 Set what kind of graticule to draw.
14233 Set graticule opacity.
14236 Set graticule flags.
14240 Draw graticule for white point.
14243 Draw graticule for black point.
14246 Draw color points short names.
14250 Set background opacity.
14252 @item lthreshold, l
14253 Set low threshold for color component not represented on X or Y axis.
14254 Values lower than this value will be ignored. Default is 0.
14255 Note this value is multiplied with actual max possible value one pixel component
14256 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14259 @item hthreshold, h
14260 Set high threshold for color component not represented on X or Y axis.
14261 Values higher than this value will be ignored. Default is 1.
14262 Note this value is multiplied with actual max possible value one pixel component
14263 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14264 is 0.9 * 255 = 230.
14266 @item colorspace, c
14267 Set what kind of colorspace to use when drawing graticule.
14276 @anchor{vidstabdetect}
14277 @section vidstabdetect
14279 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14280 @ref{vidstabtransform} for pass 2.
14282 This filter generates a file with relative translation and rotation
14283 transform information about subsequent frames, which is then used by
14284 the @ref{vidstabtransform} filter.
14286 To enable compilation of this filter you need to configure FFmpeg with
14287 @code{--enable-libvidstab}.
14289 This filter accepts the following options:
14293 Set the path to the file used to write the transforms information.
14294 Default value is @file{transforms.trf}.
14297 Set how shaky the video is and how quick the camera is. It accepts an
14298 integer in the range 1-10, a value of 1 means little shakiness, a
14299 value of 10 means strong shakiness. Default value is 5.
14302 Set the accuracy of the detection process. It must be a value in the
14303 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14304 accuracy. Default value is 15.
14307 Set stepsize of the search process. The region around minimum is
14308 scanned with 1 pixel resolution. Default value is 6.
14311 Set minimum contrast. Below this value a local measurement field is
14312 discarded. Must be a floating point value in the range 0-1. Default
14316 Set reference frame number for tripod mode.
14318 If enabled, the motion of the frames is compared to a reference frame
14319 in the filtered stream, identified by the specified number. The idea
14320 is to compensate all movements in a more-or-less static scene and keep
14321 the camera view absolutely still.
14323 If set to 0, it is disabled. The frames are counted starting from 1.
14326 Show fields and transforms in the resulting frames. It accepts an
14327 integer in the range 0-2. Default value is 0, which disables any
14331 @subsection Examples
14335 Use default values:
14341 Analyze strongly shaky movie and put the results in file
14342 @file{mytransforms.trf}:
14344 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14348 Visualize the result of internal transformations in the resulting
14351 vidstabdetect=show=1
14355 Analyze a video with medium shakiness using @command{ffmpeg}:
14357 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14361 @anchor{vidstabtransform}
14362 @section vidstabtransform
14364 Video stabilization/deshaking: pass 2 of 2,
14365 see @ref{vidstabdetect} for pass 1.
14367 Read a file with transform information for each frame and
14368 apply/compensate them. Together with the @ref{vidstabdetect}
14369 filter this can be used to deshake videos. See also
14370 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14371 the @ref{unsharp} filter, see below.
14373 To enable compilation of this filter you need to configure FFmpeg with
14374 @code{--enable-libvidstab}.
14376 @subsection Options
14380 Set path to the file used to read the transforms. Default value is
14381 @file{transforms.trf}.
14384 Set the number of frames (value*2 + 1) used for lowpass filtering the
14385 camera movements. Default value is 10.
14387 For example a number of 10 means that 21 frames are used (10 in the
14388 past and 10 in the future) to smoothen the motion in the video. A
14389 larger value leads to a smoother video, but limits the acceleration of
14390 the camera (pan/tilt movements). 0 is a special case where a static
14391 camera is simulated.
14394 Set the camera path optimization algorithm.
14396 Accepted values are:
14399 gaussian kernel low-pass filter on camera motion (default)
14401 averaging on transformations
14405 Set maximal number of pixels to translate frames. Default value is -1,
14409 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14410 value is -1, meaning no limit.
14413 Specify how to deal with borders that may be visible due to movement
14416 Available values are:
14419 keep image information from previous frame (default)
14421 fill the border black
14425 Invert transforms if set to 1. Default value is 0.
14428 Consider transforms as relative to previous frame if set to 1,
14429 absolute if set to 0. Default value is 0.
14432 Set percentage to zoom. A positive value will result in a zoom-in
14433 effect, a negative value in a zoom-out effect. Default value is 0 (no
14437 Set optimal zooming to avoid borders.
14439 Accepted values are:
14444 optimal static zoom value is determined (only very strong movements
14445 will lead to visible borders) (default)
14447 optimal adaptive zoom value is determined (no borders will be
14448 visible), see @option{zoomspeed}
14451 Note that the value given at zoom is added to the one calculated here.
14454 Set percent to zoom maximally each frame (enabled when
14455 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14459 Specify type of interpolation.
14461 Available values are:
14466 linear only horizontal
14468 linear in both directions (default)
14470 cubic in both directions (slow)
14474 Enable virtual tripod mode if set to 1, which is equivalent to
14475 @code{relative=0:smoothing=0}. Default value is 0.
14477 Use also @code{tripod} option of @ref{vidstabdetect}.
14480 Increase log verbosity if set to 1. Also the detected global motions
14481 are written to the temporary file @file{global_motions.trf}. Default
14485 @subsection Examples
14489 Use @command{ffmpeg} for a typical stabilization with default values:
14491 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14494 Note the use of the @ref{unsharp} filter which is always recommended.
14497 Zoom in a bit more and load transform data from a given file:
14499 vidstabtransform=zoom=5:input="mytransforms.trf"
14503 Smoothen the video even more:
14505 vidstabtransform=smoothing=30
14511 Flip the input video vertically.
14513 For example, to vertically flip a video with @command{ffmpeg}:
14515 ffmpeg -i in.avi -vf "vflip" out.avi
14521 Make or reverse a natural vignetting effect.
14523 The filter accepts the following options:
14527 Set lens angle expression as a number of radians.
14529 The value is clipped in the @code{[0,PI/2]} range.
14531 Default value: @code{"PI/5"}
14535 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14539 Set forward/backward mode.
14541 Available modes are:
14544 The larger the distance from the central point, the darker the image becomes.
14547 The larger the distance from the central point, the brighter the image becomes.
14548 This can be used to reverse a vignette effect, though there is no automatic
14549 detection to extract the lens @option{angle} and other settings (yet). It can
14550 also be used to create a burning effect.
14553 Default value is @samp{forward}.
14556 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14558 It accepts the following values:
14561 Evaluate expressions only once during the filter initialization.
14564 Evaluate expressions for each incoming frame. This is way slower than the
14565 @samp{init} mode since it requires all the scalers to be re-computed, but it
14566 allows advanced dynamic expressions.
14569 Default value is @samp{init}.
14572 Set dithering to reduce the circular banding effects. Default is @code{1}
14576 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14577 Setting this value to the SAR of the input will make a rectangular vignetting
14578 following the dimensions of the video.
14580 Default is @code{1/1}.
14583 @subsection Expressions
14585 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14586 following parameters.
14591 input width and height
14594 the number of input frame, starting from 0
14597 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14598 @var{TB} units, NAN if undefined
14601 frame rate of the input video, NAN if the input frame rate is unknown
14604 the PTS (Presentation TimeStamp) of the filtered video frame,
14605 expressed in seconds, NAN if undefined
14608 time base of the input video
14612 @subsection Examples
14616 Apply simple strong vignetting effect:
14622 Make a flickering vignetting:
14624 vignette='PI/4+random(1)*PI/50':eval=frame
14630 Stack input videos vertically.
14632 All streams must be of same pixel format and of same width.
14634 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14635 to create same output.
14637 The filter accept the following option:
14641 Set number of input streams. Default is 2.
14644 If set to 1, force the output to terminate when the shortest input
14645 terminates. Default value is 0.
14650 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14651 Deinterlacing Filter").
14653 Based on the process described by Martin Weston for BBC R&D, and
14654 implemented based on the de-interlace algorithm written by Jim
14655 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14656 uses filter coefficients calculated by BBC R&D.
14658 There are two sets of filter coefficients, so called "simple":
14659 and "complex". Which set of filter coefficients is used can
14660 be set by passing an optional parameter:
14664 Set the interlacing filter coefficients. Accepts one of the following values:
14668 Simple filter coefficient set.
14670 More-complex filter coefficient set.
14672 Default value is @samp{complex}.
14675 Specify which frames to deinterlace. Accept one of the following values:
14679 Deinterlace all frames,
14681 Only deinterlace frames marked as interlaced.
14684 Default value is @samp{all}.
14688 Video waveform monitor.
14690 The waveform monitor plots color component intensity. By default luminance
14691 only. Each column of the waveform corresponds to a column of pixels in the
14694 It accepts the following options:
14698 Can be either @code{row}, or @code{column}. Default is @code{column}.
14699 In row mode, the graph on the left side represents color component value 0 and
14700 the right side represents value = 255. In column mode, the top side represents
14701 color component value = 0 and bottom side represents value = 255.
14704 Set intensity. Smaller values are useful to find out how many values of the same
14705 luminance are distributed across input rows/columns.
14706 Default value is @code{0.04}. Allowed range is [0, 1].
14709 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14710 In mirrored mode, higher values will be represented on the left
14711 side for @code{row} mode and at the top for @code{column} mode. Default is
14712 @code{1} (mirrored).
14716 It accepts the following values:
14719 Presents information identical to that in the @code{parade}, except
14720 that the graphs representing color components are superimposed directly
14723 This display mode makes it easier to spot relative differences or similarities
14724 in overlapping areas of the color components that are supposed to be identical,
14725 such as neutral whites, grays, or blacks.
14728 Display separate graph for the color components side by side in
14729 @code{row} mode or one below the other in @code{column} mode.
14732 Display separate graph for the color components side by side in
14733 @code{column} mode or one below the other in @code{row} mode.
14735 Using this display mode makes it easy to spot color casts in the highlights
14736 and shadows of an image, by comparing the contours of the top and the bottom
14737 graphs of each waveform. Since whites, grays, and blacks are characterized
14738 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14739 should display three waveforms of roughly equal width/height. If not, the
14740 correction is easy to perform by making level adjustments the three waveforms.
14742 Default is @code{stack}.
14744 @item components, c
14745 Set which color components to display. Default is 1, which means only luminance
14746 or red color component if input is in RGB colorspace. If is set for example to
14747 7 it will display all 3 (if) available color components.
14752 No envelope, this is default.
14755 Instant envelope, minimum and maximum values presented in graph will be easily
14756 visible even with small @code{step} value.
14759 Hold minimum and maximum values presented in graph across time. This way you
14760 can still spot out of range values without constantly looking at waveforms.
14763 Peak and instant envelope combined together.
14769 No filtering, this is default.
14772 Luma and chroma combined together.
14775 Similar as above, but shows difference between blue and red chroma.
14778 Displays only chroma.
14781 Displays actual color value on waveform.
14784 Similar as above, but with luma showing frequency of chroma values.
14788 Set which graticule to display.
14792 Do not display graticule.
14795 Display green graticule showing legal broadcast ranges.
14799 Set graticule opacity.
14802 Set graticule flags.
14806 Draw numbers above lines. By default enabled.
14809 Draw dots instead of lines.
14813 Set scale used for displaying graticule.
14820 Default is digital.
14823 Set background opacity.
14826 @section weave, doubleweave
14828 The @code{weave} takes a field-based video input and join
14829 each two sequential fields into single frame, producing a new double
14830 height clip with half the frame rate and half the frame count.
14832 The @code{doubleweave} works same as @code{weave} but without
14833 halving frame rate and frame count.
14835 It accepts the following option:
14839 Set first field. Available values are:
14843 Set the frame as top-field-first.
14846 Set the frame as bottom-field-first.
14850 @subsection Examples
14854 Interlace video using @ref{select} and @ref{separatefields} filter:
14856 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14861 Apply the xBR high-quality magnification filter which is designed for pixel
14862 art. It follows a set of edge-detection rules, see
14863 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14865 It accepts the following option:
14869 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14870 @code{3xBR} and @code{4} for @code{4xBR}.
14871 Default is @code{3}.
14877 Deinterlace the input video ("yadif" means "yet another deinterlacing
14880 It accepts the following parameters:
14886 The interlacing mode to adopt. It accepts one of the following values:
14889 @item 0, send_frame
14890 Output one frame for each frame.
14891 @item 1, send_field
14892 Output one frame for each field.
14893 @item 2, send_frame_nospatial
14894 Like @code{send_frame}, but it skips the spatial interlacing check.
14895 @item 3, send_field_nospatial
14896 Like @code{send_field}, but it skips the spatial interlacing check.
14899 The default value is @code{send_frame}.
14902 The picture field parity assumed for the input interlaced video. It accepts one
14903 of the following values:
14907 Assume the top field is first.
14909 Assume the bottom field is first.
14911 Enable automatic detection of field parity.
14914 The default value is @code{auto}.
14915 If the interlacing is unknown or the decoder does not export this information,
14916 top field first will be assumed.
14919 Specify which frames to deinterlace. Accept one of the following
14924 Deinterlace all frames.
14925 @item 1, interlaced
14926 Only deinterlace frames marked as interlaced.
14929 The default value is @code{all}.
14934 Apply Zoom & Pan effect.
14936 This filter accepts the following options:
14940 Set the zoom expression. Default is 1.
14944 Set the x and y expression. Default is 0.
14947 Set the duration expression in number of frames.
14948 This sets for how many number of frames effect will last for
14949 single input image.
14952 Set the output image size, default is 'hd720'.
14955 Set the output frame rate, default is '25'.
14958 Each expression can contain the following constants:
14977 Output frame count.
14981 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14982 for current input frame.
14986 'x' and 'y' of last output frame of previous input frame or 0 when there was
14987 not yet such frame (first input frame).
14990 Last calculated zoom from 'z' expression for current input frame.
14993 Last calculated zoom of last output frame of previous input frame.
14996 Number of output frames for current input frame. Calculated from 'd' expression
14997 for each input frame.
15000 number of output frames created for previous input frame
15003 Rational number: input width / input height
15006 sample aspect ratio
15009 display aspect ratio
15013 @subsection Examples
15017 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
15019 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='if(gte(zoom,1.5),x,x+1/a)':y='if(gte(zoom,1.5),y,y+1)':s=640x360
15023 Zoom-in up to 1.5 and pan always at center of picture:
15025 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
15029 Same as above but without pausing:
15031 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
15036 Scale (resize) the input video, using the z.lib library:
15037 https://github.com/sekrit-twc/zimg.
15039 The zscale filter forces the output display aspect ratio to be the same
15040 as the input, by changing the output sample aspect ratio.
15042 If the input image format is different from the format requested by
15043 the next filter, the zscale filter will convert the input to the
15046 @subsection Options
15047 The filter accepts the following options.
15052 Set the output video dimension expression. Default value is the input
15055 If the @var{width} or @var{w} is 0, the input width is used for the output.
15056 If the @var{height} or @var{h} is 0, the input height is used for the output.
15058 If one of the values is -1, the zscale filter will use a value that
15059 maintains the aspect ratio of the input image, calculated from the
15060 other specified dimension. If both of them are -1, the input size is
15063 If one of the values is -n with n > 1, the zscale filter will also use a value
15064 that maintains the aspect ratio of the input image, calculated from the other
15065 specified dimension. After that it will, however, make sure that the calculated
15066 dimension is divisible by n and adjust the value if necessary.
15068 See below for the list of accepted constants for use in the dimension
15072 Set the video size. For the syntax of this option, check the
15073 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15076 Set the dither type.
15078 Possible values are:
15083 @item error_diffusion
15089 Set the resize filter type.
15091 Possible values are:
15101 Default is bilinear.
15104 Set the color range.
15106 Possible values are:
15113 Default is same as input.
15116 Set the color primaries.
15118 Possible values are:
15128 Default is same as input.
15131 Set the transfer characteristics.
15133 Possible values are:
15147 Default is same as input.
15150 Set the colorspace matrix.
15152 Possible value are:
15163 Default is same as input.
15166 Set the input color range.
15168 Possible values are:
15175 Default is same as input.
15177 @item primariesin, pin
15178 Set the input color primaries.
15180 Possible values are:
15190 Default is same as input.
15192 @item transferin, tin
15193 Set the input transfer characteristics.
15195 Possible values are:
15206 Default is same as input.
15208 @item matrixin, min
15209 Set the input colorspace matrix.
15211 Possible value are:
15223 Set the output chroma location.
15225 Possible values are:
15236 @item chromalin, cin
15237 Set the input chroma location.
15239 Possible values are:
15251 Set the nominal peak luminance.
15254 The values of the @option{w} and @option{h} options are expressions
15255 containing the following constants:
15260 The input width and height
15264 These are the same as @var{in_w} and @var{in_h}.
15268 The output (scaled) width and height
15272 These are the same as @var{out_w} and @var{out_h}
15275 The same as @var{iw} / @var{ih}
15278 input sample aspect ratio
15281 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15285 horizontal and vertical input chroma subsample values. For example for the
15286 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15290 horizontal and vertical output chroma subsample values. For example for the
15291 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15297 @c man end VIDEO FILTERS
15299 @chapter Video Sources
15300 @c man begin VIDEO SOURCES
15302 Below is a description of the currently available video sources.
15306 Buffer video frames, and make them available to the filter chain.
15308 This source is mainly intended for a programmatic use, in particular
15309 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15311 It accepts the following parameters:
15316 Specify the size (width and height) of the buffered video frames. For the
15317 syntax of this option, check the
15318 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15321 The input video width.
15324 The input video height.
15327 A string representing the pixel format of the buffered video frames.
15328 It may be a number corresponding to a pixel format, or a pixel format
15332 Specify the timebase assumed by the timestamps of the buffered frames.
15335 Specify the frame rate expected for the video stream.
15337 @item pixel_aspect, sar
15338 The sample (pixel) aspect ratio of the input video.
15341 Specify the optional parameters to be used for the scale filter which
15342 is automatically inserted when an input change is detected in the
15343 input size or format.
15345 @item hw_frames_ctx
15346 When using a hardware pixel format, this should be a reference to an
15347 AVHWFramesContext describing input frames.
15352 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15355 will instruct the source to accept video frames with size 320x240 and
15356 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15357 square pixels (1:1 sample aspect ratio).
15358 Since the pixel format with name "yuv410p" corresponds to the number 6
15359 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15360 this example corresponds to:
15362 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15365 Alternatively, the options can be specified as a flat string, but this
15366 syntax is deprecated:
15368 @var{width}:@var{height}:@var{pix_fmt}:@var{time_base.num}:@var{time_base.den}:@var{pixel_aspect.num}:@var{pixel_aspect.den}[:@var{sws_param}]
15372 Create a pattern generated by an elementary cellular automaton.
15374 The initial state of the cellular automaton can be defined through the
15375 @option{filename} and @option{pattern} options. If such options are
15376 not specified an initial state is created randomly.
15378 At each new frame a new row in the video is filled with the result of
15379 the cellular automaton next generation. The behavior when the whole
15380 frame is filled is defined by the @option{scroll} option.
15382 This source accepts the following options:
15386 Read the initial cellular automaton state, i.e. the starting row, from
15387 the specified file.
15388 In the file, each non-whitespace character is considered an alive
15389 cell, a newline will terminate the row, and further characters in the
15390 file will be ignored.
15393 Read the initial cellular automaton state, i.e. the starting row, from
15394 the specified string.
15396 Each non-whitespace character in the string is considered an alive
15397 cell, a newline will terminate the row, and further characters in the
15398 string will be ignored.
15401 Set the video rate, that is the number of frames generated per second.
15404 @item random_fill_ratio, ratio
15405 Set the random fill ratio for the initial cellular automaton row. It
15406 is a floating point number value ranging from 0 to 1, defaults to
15409 This option is ignored when a file or a pattern is specified.
15411 @item random_seed, seed
15412 Set the seed for filling randomly the initial row, must be an integer
15413 included between 0 and UINT32_MAX. If not specified, or if explicitly
15414 set to -1, the filter will try to use a good random seed on a best
15418 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15419 Default value is 110.
15422 Set the size of the output video. For the syntax of this option, check the
15423 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15425 If @option{filename} or @option{pattern} is specified, the size is set
15426 by default to the width of the specified initial state row, and the
15427 height is set to @var{width} * PHI.
15429 If @option{size} is set, it must contain the width of the specified
15430 pattern string, and the specified pattern will be centered in the
15433 If a filename or a pattern string is not specified, the size value
15434 defaults to "320x518" (used for a randomly generated initial state).
15437 If set to 1, scroll the output upward when all the rows in the output
15438 have been already filled. If set to 0, the new generated row will be
15439 written over the top row just after the bottom row is filled.
15442 @item start_full, full
15443 If set to 1, completely fill the output with generated rows before
15444 outputting the first frame.
15445 This is the default behavior, for disabling set the value to 0.
15448 If set to 1, stitch the left and right row edges together.
15449 This is the default behavior, for disabling set the value to 0.
15452 @subsection Examples
15456 Read the initial state from @file{pattern}, and specify an output of
15459 cellauto=f=pattern:s=200x400
15463 Generate a random initial row with a width of 200 cells, with a fill
15466 cellauto=ratio=2/3:s=200x200
15470 Create a pattern generated by rule 18 starting by a single alive cell
15471 centered on an initial row with width 100:
15473 cellauto=p=@@:s=100x400:full=0:rule=18
15477 Specify a more elaborated initial pattern:
15479 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15484 @anchor{coreimagesrc}
15485 @section coreimagesrc
15486 Video source generated on GPU using Apple's CoreImage API on OSX.
15488 This video source is a specialized version of the @ref{coreimage} video filter.
15489 Use a core image generator at the beginning of the applied filterchain to
15490 generate the content.
15492 The coreimagesrc video source accepts the following options:
15494 @item list_generators
15495 List all available generators along with all their respective options as well as
15496 possible minimum and maximum values along with the default values.
15498 list_generators=true
15502 Specify the size of the sourced video. For the syntax of this option, check the
15503 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15504 The default value is @code{320x240}.
15507 Specify the frame rate of the sourced video, as the number of frames
15508 generated per second. It has to be a string in the format
15509 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15510 number or a valid video frame rate abbreviation. The default value is
15514 Set the sample aspect ratio of the sourced video.
15517 Set the duration of the sourced video. See
15518 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15519 for the accepted syntax.
15521 If not specified, or the expressed duration is negative, the video is
15522 supposed to be generated forever.
15525 Additionally, all options of the @ref{coreimage} video filter are accepted.
15526 A complete filterchain can be used for further processing of the
15527 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15528 and examples for details.
15530 @subsection Examples
15535 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15536 given as complete and escaped command-line for Apple's standard bash shell:
15538 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15540 This example is equivalent to the QRCode example of @ref{coreimage} without the
15541 need for a nullsrc video source.
15545 @section mandelbrot
15547 Generate a Mandelbrot set fractal, and progressively zoom towards the
15548 point specified with @var{start_x} and @var{start_y}.
15550 This source accepts the following options:
15555 Set the terminal pts value. Default value is 400.
15558 Set the terminal scale value.
15559 Must be a floating point value. Default value is 0.3.
15562 Set the inner coloring mode, that is the algorithm used to draw the
15563 Mandelbrot fractal internal region.
15565 It shall assume one of the following values:
15570 Show time until convergence.
15572 Set color based on point closest to the origin of the iterations.
15577 Default value is @var{mincol}.
15580 Set the bailout value. Default value is 10.0.
15583 Set the maximum of iterations performed by the rendering
15584 algorithm. Default value is 7189.
15587 Set outer coloring mode.
15588 It shall assume one of following values:
15590 @item iteration_count
15591 Set iteration cound mode.
15592 @item normalized_iteration_count
15593 set normalized iteration count mode.
15595 Default value is @var{normalized_iteration_count}.
15598 Set frame rate, expressed as number of frames per second. Default
15602 Set frame size. For the syntax of this option, check the "Video
15603 size" section in the ffmpeg-utils manual. Default value is "640x480".
15606 Set the initial scale value. Default value is 3.0.
15609 Set the initial x position. Must be a floating point value between
15610 -100 and 100. Default value is -0.743643887037158704752191506114774.
15613 Set the initial y position. Must be a floating point value between
15614 -100 and 100. Default value is -0.131825904205311970493132056385139.
15619 Generate various test patterns, as generated by the MPlayer test filter.
15621 The size of the generated video is fixed, and is 256x256.
15622 This source is useful in particular for testing encoding features.
15624 This source accepts the following options:
15629 Specify the frame rate of the sourced video, as the number of frames
15630 generated per second. It has to be a string in the format
15631 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15632 number or a valid video frame rate abbreviation. The default value is
15636 Set the duration of the sourced video. See
15637 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15638 for the accepted syntax.
15640 If not specified, or the expressed duration is negative, the video is
15641 supposed to be generated forever.
15645 Set the number or the name of the test to perform. Supported tests are:
15661 Default value is "all", which will cycle through the list of all tests.
15666 mptestsrc=t=dc_luma
15669 will generate a "dc_luma" test pattern.
15671 @section frei0r_src
15673 Provide a frei0r source.
15675 To enable compilation of this filter you need to install the frei0r
15676 header and configure FFmpeg with @code{--enable-frei0r}.
15678 This source accepts the following parameters:
15683 The size of the video to generate. For the syntax of this option, check the
15684 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15687 The framerate of the generated video. It may be a string of the form
15688 @var{num}/@var{den} or a frame rate abbreviation.
15691 The name to the frei0r source to load. For more information regarding frei0r and
15692 how to set the parameters, read the @ref{frei0r} section in the video filters
15695 @item filter_params
15696 A '|'-separated list of parameters to pass to the frei0r source.
15700 For example, to generate a frei0r partik0l source with size 200x200
15701 and frame rate 10 which is overlaid on the overlay filter main input:
15703 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15708 Generate a life pattern.
15710 This source is based on a generalization of John Conway's life game.
15712 The sourced input represents a life grid, each pixel represents a cell
15713 which can be in one of two possible states, alive or dead. Every cell
15714 interacts with its eight neighbours, which are the cells that are
15715 horizontally, vertically, or diagonally adjacent.
15717 At each interaction the grid evolves according to the adopted rule,
15718 which specifies the number of neighbor alive cells which will make a
15719 cell stay alive or born. The @option{rule} option allows one to specify
15722 This source accepts the following options:
15726 Set the file from which to read the initial grid state. In the file,
15727 each non-whitespace character is considered an alive cell, and newline
15728 is used to delimit the end of each row.
15730 If this option is not specified, the initial grid is generated
15734 Set the video rate, that is the number of frames generated per second.
15737 @item random_fill_ratio, ratio
15738 Set the random fill ratio for the initial random grid. It is a
15739 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15740 It is ignored when a file is specified.
15742 @item random_seed, seed
15743 Set the seed for filling the initial random grid, must be an integer
15744 included between 0 and UINT32_MAX. If not specified, or if explicitly
15745 set to -1, the filter will try to use a good random seed on a best
15751 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15752 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15753 @var{NS} specifies the number of alive neighbor cells which make a
15754 live cell stay alive, and @var{NB} the number of alive neighbor cells
15755 which make a dead cell to become alive (i.e. to "born").
15756 "s" and "b" can be used in place of "S" and "B", respectively.
15758 Alternatively a rule can be specified by an 18-bits integer. The 9
15759 high order bits are used to encode the next cell state if it is alive
15760 for each number of neighbor alive cells, the low order bits specify
15761 the rule for "borning" new cells. Higher order bits encode for an
15762 higher number of neighbor cells.
15763 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15764 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15766 Default value is "S23/B3", which is the original Conway's game of life
15767 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15768 cells, and will born a new cell if there are three alive cells around
15772 Set the size of the output video. For the syntax of this option, check the
15773 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15775 If @option{filename} is specified, the size is set by default to the
15776 same size of the input file. If @option{size} is set, it must contain
15777 the size specified in the input file, and the initial grid defined in
15778 that file is centered in the larger resulting area.
15780 If a filename is not specified, the size value defaults to "320x240"
15781 (used for a randomly generated initial grid).
15784 If set to 1, stitch the left and right grid edges together, and the
15785 top and bottom edges also. Defaults to 1.
15788 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15789 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15790 value from 0 to 255.
15793 Set the color of living (or new born) cells.
15796 Set the color of dead cells. If @option{mold} is set, this is the first color
15797 used to represent a dead cell.
15800 Set mold color, for definitely dead and moldy cells.
15802 For the syntax of these 3 color options, check the "Color" section in the
15803 ffmpeg-utils manual.
15806 @subsection Examples
15810 Read a grid from @file{pattern}, and center it on a grid of size
15813 life=f=pattern:s=300x300
15817 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15819 life=ratio=2/3:s=200x200
15823 Specify a custom rule for evolving a randomly generated grid:
15829 Full example with slow death effect (mold) using @command{ffplay}:
15831 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15838 @anchor{haldclutsrc}
15840 @anchor{rgbtestsrc}
15842 @anchor{smptehdbars}
15845 @anchor{yuvtestsrc}
15846 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15848 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15850 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15852 The @code{color} source provides an uniformly colored input.
15854 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15855 @ref{haldclut} filter.
15857 The @code{nullsrc} source returns unprocessed video frames. It is
15858 mainly useful to be employed in analysis / debugging tools, or as the
15859 source for filters which ignore the input data.
15861 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15862 detecting RGB vs BGR issues. You should see a red, green and blue
15863 stripe from top to bottom.
15865 The @code{smptebars} source generates a color bars pattern, based on
15866 the SMPTE Engineering Guideline EG 1-1990.
15868 The @code{smptehdbars} source generates a color bars pattern, based on
15869 the SMPTE RP 219-2002.
15871 The @code{testsrc} source generates a test video pattern, showing a
15872 color pattern, a scrolling gradient and a timestamp. This is mainly
15873 intended for testing purposes.
15875 The @code{testsrc2} source is similar to testsrc, but supports more
15876 pixel formats instead of just @code{rgb24}. This allows using it as an
15877 input for other tests without requiring a format conversion.
15879 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15880 see a y, cb and cr stripe from top to bottom.
15882 The sources accept the following parameters:
15887 Specify the color of the source, only available in the @code{color}
15888 source. For the syntax of this option, check the "Color" section in the
15889 ffmpeg-utils manual.
15892 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15893 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15894 pixels to be used as identity matrix for 3D lookup tables. Each component is
15895 coded on a @code{1/(N*N)} scale.
15898 Specify the size of the sourced video. For the syntax of this option, check the
15899 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15900 The default value is @code{320x240}.
15902 This option is not available with the @code{haldclutsrc} filter.
15905 Specify the frame rate of the sourced video, as the number of frames
15906 generated per second. It has to be a string in the format
15907 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15908 number or a valid video frame rate abbreviation. The default value is
15912 Set the sample aspect ratio of the sourced video.
15915 Set the duration of the sourced video. See
15916 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15917 for the accepted syntax.
15919 If not specified, or the expressed duration is negative, the video is
15920 supposed to be generated forever.
15923 Set the number of decimals to show in the timestamp, only available in the
15924 @code{testsrc} source.
15926 The displayed timestamp value will correspond to the original
15927 timestamp value multiplied by the power of 10 of the specified
15928 value. Default value is 0.
15931 For example the following:
15933 testsrc=duration=5.3:size=qcif:rate=10
15936 will generate a video with a duration of 5.3 seconds, with size
15937 176x144 and a frame rate of 10 frames per second.
15939 The following graph description will generate a red source
15940 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15943 color=c=red@@0.2:s=qcif:r=10
15946 If the input content is to be ignored, @code{nullsrc} can be used. The
15947 following command generates noise in the luminance plane by employing
15948 the @code{geq} filter:
15950 nullsrc=s=256x256, geq=random(1)*255:128:128
15953 @subsection Commands
15955 The @code{color} source supports the following commands:
15959 Set the color of the created image. Accepts the same syntax of the
15960 corresponding @option{color} option.
15963 @c man end VIDEO SOURCES
15965 @chapter Video Sinks
15966 @c man begin VIDEO SINKS
15968 Below is a description of the currently available video sinks.
15970 @section buffersink
15972 Buffer video frames, and make them available to the end of the filter
15975 This sink is mainly intended for programmatic use, in particular
15976 through the interface defined in @file{libavfilter/buffersink.h}
15977 or the options system.
15979 It accepts a pointer to an AVBufferSinkContext structure, which
15980 defines the incoming buffers' formats, to be passed as the opaque
15981 parameter to @code{avfilter_init_filter} for initialization.
15985 Null video sink: do absolutely nothing with the input video. It is
15986 mainly useful as a template and for use in analysis / debugging
15989 @c man end VIDEO SINKS
15991 @chapter Multimedia Filters
15992 @c man begin MULTIMEDIA FILTERS
15994 Below is a description of the currently available multimedia filters.
15998 Convert input audio to a video output, displaying the audio bit scope.
16000 The filter accepts the following options:
16004 Set frame rate, expressed as number of frames per second. Default
16008 Specify the video size for the output. For the syntax of this option, check the
16009 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16010 Default value is @code{1024x256}.
16013 Specify list of colors separated by space or by '|' which will be used to
16014 draw channels. Unrecognized or missing colors will be replaced
16018 @section ahistogram
16020 Convert input audio to a video output, displaying the volume histogram.
16022 The filter accepts the following options:
16026 Specify how histogram is calculated.
16028 It accepts the following values:
16031 Use single histogram for all channels.
16033 Use separate histogram for each channel.
16035 Default is @code{single}.
16038 Set frame rate, expressed as number of frames per second. Default
16042 Specify the video size for the output. For the syntax of this option, check the
16043 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16044 Default value is @code{hd720}.
16049 It accepts the following values:
16060 reverse logarithmic
16062 Default is @code{log}.
16065 Set amplitude scale.
16067 It accepts the following values:
16074 Default is @code{log}.
16077 Set how much frames to accumulate in histogram.
16078 Defauls is 1. Setting this to -1 accumulates all frames.
16081 Set histogram ratio of window height.
16084 Set sonogram sliding.
16086 It accepts the following values:
16089 replace old rows with new ones.
16091 scroll from top to bottom.
16093 Default is @code{replace}.
16096 @section aphasemeter
16098 Convert input audio to a video output, displaying the audio phase.
16100 The filter accepts the following options:
16104 Set the output frame rate. Default value is @code{25}.
16107 Set the video size for the output. For the syntax of this option, check the
16108 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16109 Default value is @code{800x400}.
16114 Specify the red, green, blue contrast. Default values are @code{2},
16115 @code{7} and @code{1}.
16116 Allowed range is @code{[0, 255]}.
16119 Set color which will be used for drawing median phase. If color is
16120 @code{none} which is default, no median phase value will be drawn.
16123 Enable video output. Default is enabled.
16126 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
16127 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
16128 The @code{-1} means left and right channels are completely out of phase and
16129 @code{1} means channels are in phase.
16131 @section avectorscope
16133 Convert input audio to a video output, representing the audio vector
16136 The filter is used to measure the difference between channels of stereo
16137 audio stream. A monoaural signal, consisting of identical left and right
16138 signal, results in straight vertical line. Any stereo separation is visible
16139 as a deviation from this line, creating a Lissajous figure.
16140 If the straight (or deviation from it) but horizontal line appears this
16141 indicates that the left and right channels are out of phase.
16143 The filter accepts the following options:
16147 Set the vectorscope mode.
16149 Available values are:
16152 Lissajous rotated by 45 degrees.
16155 Same as above but not rotated.
16158 Shape resembling half of circle.
16161 Default value is @samp{lissajous}.
16164 Set the video size for the output. For the syntax of this option, check the
16165 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16166 Default value is @code{400x400}.
16169 Set the output frame rate. Default value is @code{25}.
16175 Specify the red, green, blue and alpha contrast. Default values are @code{40},
16176 @code{160}, @code{80} and @code{255}.
16177 Allowed range is @code{[0, 255]}.
16183 Specify the red, green, blue and alpha fade. Default values are @code{15},
16184 @code{10}, @code{5} and @code{5}.
16185 Allowed range is @code{[0, 255]}.
16188 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
16191 Set the vectorscope drawing mode.
16193 Available values are:
16196 Draw dot for each sample.
16199 Draw line between previous and current sample.
16202 Default value is @samp{dot}.
16205 Specify amplitude scale of audio samples.
16207 Available values are:
16224 @subsection Examples
16228 Complete example using @command{ffplay}:
16230 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16231 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16235 @section bench, abench
16237 Benchmark part of a filtergraph.
16239 The filter accepts the following options:
16243 Start or stop a timer.
16245 Available values are:
16248 Get the current time, set it as frame metadata (using the key
16249 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
16252 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16253 the input frame metadata to get the time difference. Time difference, average,
16254 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16255 @code{min}) are then printed. The timestamps are expressed in seconds.
16259 @subsection Examples
16263 Benchmark @ref{selectivecolor} filter:
16265 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16271 Concatenate audio and video streams, joining them together one after the
16274 The filter works on segments of synchronized video and audio streams. All
16275 segments must have the same number of streams of each type, and that will
16276 also be the number of streams at output.
16278 The filter accepts the following options:
16283 Set the number of segments. Default is 2.
16286 Set the number of output video streams, that is also the number of video
16287 streams in each segment. Default is 1.
16290 Set the number of output audio streams, that is also the number of audio
16291 streams in each segment. Default is 0.
16294 Activate unsafe mode: do not fail if segments have a different format.
16298 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16299 @var{a} audio outputs.
16301 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16302 segment, in the same order as the outputs, then the inputs for the second
16305 Related streams do not always have exactly the same duration, for various
16306 reasons including codec frame size or sloppy authoring. For that reason,
16307 related synchronized streams (e.g. a video and its audio track) should be
16308 concatenated at once. The concat filter will use the duration of the longest
16309 stream in each segment (except the last one), and if necessary pad shorter
16310 audio streams with silence.
16312 For this filter to work correctly, all segments must start at timestamp 0.
16314 All corresponding streams must have the same parameters in all segments; the
16315 filtering system will automatically select a common pixel format for video
16316 streams, and a common sample format, sample rate and channel layout for
16317 audio streams, but other settings, such as resolution, must be converted
16318 explicitly by the user.
16320 Different frame rates are acceptable but will result in variable frame rate
16321 at output; be sure to configure the output file to handle it.
16323 @subsection Examples
16327 Concatenate an opening, an episode and an ending, all in bilingual version
16328 (video in stream 0, audio in streams 1 and 2):
16330 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16331 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16332 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16333 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16337 Concatenate two parts, handling audio and video separately, using the
16338 (a)movie sources, and adjusting the resolution:
16340 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16341 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16342 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16344 Note that a desync will happen at the stitch if the audio and video streams
16345 do not have exactly the same duration in the first file.
16349 @section drawgraph, adrawgraph
16351 Draw a graph using input video or audio metadata.
16353 It accepts the following parameters:
16357 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16360 Set 1st foreground color expression.
16363 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16366 Set 2nd foreground color expression.
16369 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16372 Set 3rd foreground color expression.
16375 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16378 Set 4th foreground color expression.
16381 Set minimal value of metadata value.
16384 Set maximal value of metadata value.
16387 Set graph background color. Default is white.
16392 Available values for mode is:
16399 Default is @code{line}.
16404 Available values for slide is:
16407 Draw new frame when right border is reached.
16410 Replace old columns with new ones.
16413 Scroll from right to left.
16416 Scroll from left to right.
16419 Draw single picture.
16422 Default is @code{frame}.
16425 Set size of graph video. For the syntax of this option, check the
16426 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16427 The default value is @code{900x256}.
16429 The foreground color expressions can use the following variables:
16432 Minimal value of metadata value.
16435 Maximal value of metadata value.
16438 Current metadata key value.
16441 The color is defined as 0xAABBGGRR.
16444 Example using metadata from @ref{signalstats} filter:
16446 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16449 Example using metadata from @ref{ebur128} filter:
16451 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16457 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16458 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16459 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16460 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16462 The filter also has a video output (see the @var{video} option) with a real
16463 time graph to observe the loudness evolution. The graphic contains the logged
16464 message mentioned above, so it is not printed anymore when this option is set,
16465 unless the verbose logging is set. The main graphing area contains the
16466 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16467 the momentary loudness (400 milliseconds).
16469 More information about the Loudness Recommendation EBU R128 on
16470 @url{http://tech.ebu.ch/loudness}.
16472 The filter accepts the following options:
16477 Activate the video output. The audio stream is passed unchanged whether this
16478 option is set or no. The video stream will be the first output stream if
16479 activated. Default is @code{0}.
16482 Set the video size. This option is for video only. For the syntax of this
16484 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16485 Default and minimum resolution is @code{640x480}.
16488 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16489 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16490 other integer value between this range is allowed.
16493 Set metadata injection. If set to @code{1}, the audio input will be segmented
16494 into 100ms output frames, each of them containing various loudness information
16495 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16497 Default is @code{0}.
16500 Force the frame logging level.
16502 Available values are:
16505 information logging level
16507 verbose logging level
16510 By default, the logging level is set to @var{info}. If the @option{video} or
16511 the @option{metadata} options are set, it switches to @var{verbose}.
16516 Available modes can be cumulated (the option is a @code{flag} type). Possible
16520 Disable any peak mode (default).
16522 Enable sample-peak mode.
16524 Simple peak mode looking for the higher sample value. It logs a message
16525 for sample-peak (identified by @code{SPK}).
16527 Enable true-peak mode.
16529 If enabled, the peak lookup is done on an over-sampled version of the input
16530 stream for better peak accuracy. It logs a message for true-peak.
16531 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16532 This mode requires a build with @code{libswresample}.
16536 Treat mono input files as "dual mono". If a mono file is intended for playback
16537 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16538 If set to @code{true}, this option will compensate for this effect.
16539 Multi-channel input files are not affected by this option.
16542 Set a specific pan law to be used for the measurement of dual mono files.
16543 This parameter is optional, and has a default value of -3.01dB.
16546 @subsection Examples
16550 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16552 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16556 Run an analysis with @command{ffmpeg}:
16558 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16562 @section interleave, ainterleave
16564 Temporally interleave frames from several inputs.
16566 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16568 These filters read frames from several inputs and send the oldest
16569 queued frame to the output.
16571 Input streams must have well defined, monotonically increasing frame
16574 In order to submit one frame to output, these filters need to enqueue
16575 at least one frame for each input, so they cannot work in case one
16576 input is not yet terminated and will not receive incoming frames.
16578 For example consider the case when one input is a @code{select} filter
16579 which always drops input frames. The @code{interleave} filter will keep
16580 reading from that input, but it will never be able to send new frames
16581 to output until the input sends an end-of-stream signal.
16583 Also, depending on inputs synchronization, the filters will drop
16584 frames in case one input receives more frames than the other ones, and
16585 the queue is already filled.
16587 These filters accept the following options:
16591 Set the number of different inputs, it is 2 by default.
16594 @subsection Examples
16598 Interleave frames belonging to different streams using @command{ffmpeg}:
16600 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16604 Add flickering blur effect:
16606 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16610 @section metadata, ametadata
16612 Manipulate frame metadata.
16614 This filter accepts the following options:
16618 Set mode of operation of the filter.
16620 Can be one of the following:
16624 If both @code{value} and @code{key} is set, select frames
16625 which have such metadata. If only @code{key} is set, select
16626 every frame that has such key in metadata.
16629 Add new metadata @code{key} and @code{value}. If key is already available
16633 Modify value of already present key.
16636 If @code{value} is set, delete only keys that have such value.
16637 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16641 Print key and its value if metadata was found. If @code{key} is not set print all
16642 metadata values available in frame.
16646 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16649 Set metadata value which will be used. This option is mandatory for
16650 @code{modify} and @code{add} mode.
16653 Which function to use when comparing metadata value and @code{value}.
16655 Can be one of following:
16659 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16662 Values are interpreted as strings, returns true if metadata value starts with
16663 the @code{value} option string.
16666 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16669 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16672 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16675 Values are interpreted as floats, returns true if expression from option @code{expr}
16680 Set expression which is used when @code{function} is set to @code{expr}.
16681 The expression is evaluated through the eval API and can contain the following
16686 Float representation of @code{value} from metadata key.
16689 Float representation of @code{value} as supplied by user in @code{value} option.
16693 If specified in @code{print} mode, output is written to the named file. Instead of
16694 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16695 for standard output. If @code{file} option is not set, output is written to the log
16696 with AV_LOG_INFO loglevel.
16700 @subsection Examples
16704 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16707 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16710 Print silencedetect output to file @file{metadata.txt}.
16712 silencedetect,ametadata=mode=print:file=metadata.txt
16715 Direct all metadata to a pipe with file descriptor 4.
16717 metadata=mode=print:file='pipe\:4'
16721 @section perms, aperms
16723 Set read/write permissions for the output frames.
16725 These filters are mainly aimed at developers to test direct path in the
16726 following filter in the filtergraph.
16728 The filters accept the following options:
16732 Select the permissions mode.
16734 It accepts the following values:
16737 Do nothing. This is the default.
16739 Set all the output frames read-only.
16741 Set all the output frames directly writable.
16743 Make the frame read-only if writable, and writable if read-only.
16745 Set each output frame read-only or writable randomly.
16749 Set the seed for the @var{random} mode, must be an integer included between
16750 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16751 @code{-1}, the filter will try to use a good random seed on a best effort
16755 Note: in case of auto-inserted filter between the permission filter and the
16756 following one, the permission might not be received as expected in that
16757 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16758 perms/aperms filter can avoid this problem.
16760 @section realtime, arealtime
16762 Slow down filtering to match real time approximatively.
16764 These filters will pause the filtering for a variable amount of time to
16765 match the output rate with the input timestamps.
16766 They are similar to the @option{re} option to @code{ffmpeg}.
16768 They accept the following options:
16772 Time limit for the pauses. Any pause longer than that will be considered
16773 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16777 @section select, aselect
16779 Select frames to pass in output.
16781 This filter accepts the following options:
16786 Set expression, which is evaluated for each input frame.
16788 If the expression is evaluated to zero, the frame is discarded.
16790 If the evaluation result is negative or NaN, the frame is sent to the
16791 first output; otherwise it is sent to the output with index
16792 @code{ceil(val)-1}, assuming that the input index starts from 0.
16794 For example a value of @code{1.2} corresponds to the output with index
16795 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16798 Set the number of outputs. The output to which to send the selected
16799 frame is based on the result of the evaluation. Default value is 1.
16802 The expression can contain the following constants:
16806 The (sequential) number of the filtered frame, starting from 0.
16809 The (sequential) number of the selected frame, starting from 0.
16811 @item prev_selected_n
16812 The sequential number of the last selected frame. It's NAN if undefined.
16815 The timebase of the input timestamps.
16818 The PTS (Presentation TimeStamp) of the filtered video frame,
16819 expressed in @var{TB} units. It's NAN if undefined.
16822 The PTS of the filtered video frame,
16823 expressed in seconds. It's NAN if undefined.
16826 The PTS of the previously filtered video frame. It's NAN if undefined.
16828 @item prev_selected_pts
16829 The PTS of the last previously filtered video frame. It's NAN if undefined.
16831 @item prev_selected_t
16832 The PTS of the last previously selected video frame. It's NAN if undefined.
16835 The PTS of the first video frame in the video. It's NAN if undefined.
16838 The time of the first video frame in the video. It's NAN if undefined.
16840 @item pict_type @emph{(video only)}
16841 The type of the filtered frame. It can assume one of the following
16853 @item interlace_type @emph{(video only)}
16854 The frame interlace type. It can assume one of the following values:
16857 The frame is progressive (not interlaced).
16859 The frame is top-field-first.
16861 The frame is bottom-field-first.
16864 @item consumed_sample_n @emph{(audio only)}
16865 the number of selected samples before the current frame
16867 @item samples_n @emph{(audio only)}
16868 the number of samples in the current frame
16870 @item sample_rate @emph{(audio only)}
16871 the input sample rate
16874 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16877 the position in the file of the filtered frame, -1 if the information
16878 is not available (e.g. for synthetic video)
16880 @item scene @emph{(video only)}
16881 value between 0 and 1 to indicate a new scene; a low value reflects a low
16882 probability for the current frame to introduce a new scene, while a higher
16883 value means the current frame is more likely to be one (see the example below)
16885 @item concatdec_select
16886 The concat demuxer can select only part of a concat input file by setting an
16887 inpoint and an outpoint, but the output packets may not be entirely contained
16888 in the selected interval. By using this variable, it is possible to skip frames
16889 generated by the concat demuxer which are not exactly contained in the selected
16892 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16893 and the @var{lavf.concat.duration} packet metadata values which are also
16894 present in the decoded frames.
16896 The @var{concatdec_select} variable is -1 if the frame pts is at least
16897 start_time and either the duration metadata is missing or the frame pts is less
16898 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16901 That basically means that an input frame is selected if its pts is within the
16902 interval set by the concat demuxer.
16906 The default value of the select expression is "1".
16908 @subsection Examples
16912 Select all frames in input:
16917 The example above is the same as:
16929 Select only I-frames:
16931 select='eq(pict_type\,I)'
16935 Select one frame every 100:
16937 select='not(mod(n\,100))'
16941 Select only frames contained in the 10-20 time interval:
16943 select=between(t\,10\,20)
16947 Select only I-frames contained in the 10-20 time interval:
16949 select=between(t\,10\,20)*eq(pict_type\,I)
16953 Select frames with a minimum distance of 10 seconds:
16955 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16959 Use aselect to select only audio frames with samples number > 100:
16961 aselect='gt(samples_n\,100)'
16965 Create a mosaic of the first scenes:
16967 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16970 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16974 Send even and odd frames to separate outputs, and compose them:
16976 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16980 Select useful frames from an ffconcat file which is using inpoints and
16981 outpoints but where the source files are not intra frame only.
16983 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16987 @section sendcmd, asendcmd
16989 Send commands to filters in the filtergraph.
16991 These filters read commands to be sent to other filters in the
16994 @code{sendcmd} must be inserted between two video filters,
16995 @code{asendcmd} must be inserted between two audio filters, but apart
16996 from that they act the same way.
16998 The specification of commands can be provided in the filter arguments
16999 with the @var{commands} option, or in a file specified by the
17000 @var{filename} option.
17002 These filters accept the following options:
17005 Set the commands to be read and sent to the other filters.
17007 Set the filename of the commands to be read and sent to the other
17011 @subsection Commands syntax
17013 A commands description consists of a sequence of interval
17014 specifications, comprising a list of commands to be executed when a
17015 particular event related to that interval occurs. The occurring event
17016 is typically the current frame time entering or leaving a given time
17019 An interval is specified by the following syntax:
17021 @var{START}[-@var{END}] @var{COMMANDS};
17024 The time interval is specified by the @var{START} and @var{END} times.
17025 @var{END} is optional and defaults to the maximum time.
17027 The current frame time is considered within the specified interval if
17028 it is included in the interval [@var{START}, @var{END}), that is when
17029 the time is greater or equal to @var{START} and is lesser than
17032 @var{COMMANDS} consists of a sequence of one or more command
17033 specifications, separated by ",", relating to that interval. The
17034 syntax of a command specification is given by:
17036 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
17039 @var{FLAGS} is optional and specifies the type of events relating to
17040 the time interval which enable sending the specified command, and must
17041 be a non-null sequence of identifier flags separated by "+" or "|" and
17042 enclosed between "[" and "]".
17044 The following flags are recognized:
17047 The command is sent when the current frame timestamp enters the
17048 specified interval. In other words, the command is sent when the
17049 previous frame timestamp was not in the given interval, and the
17053 The command is sent when the current frame timestamp leaves the
17054 specified interval. In other words, the command is sent when the
17055 previous frame timestamp was in the given interval, and the
17059 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
17062 @var{TARGET} specifies the target of the command, usually the name of
17063 the filter class or a specific filter instance name.
17065 @var{COMMAND} specifies the name of the command for the target filter.
17067 @var{ARG} is optional and specifies the optional list of argument for
17068 the given @var{COMMAND}.
17070 Between one interval specification and another, whitespaces, or
17071 sequences of characters starting with @code{#} until the end of line,
17072 are ignored and can be used to annotate comments.
17074 A simplified BNF description of the commands specification syntax
17077 @var{COMMAND_FLAG} ::= "enter" | "leave"
17078 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
17079 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
17080 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
17081 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
17082 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
17085 @subsection Examples
17089 Specify audio tempo change at second 4:
17091 asendcmd=c='4.0 atempo tempo 1.5',atempo
17095 Specify a list of drawtext and hue commands in a file.
17097 # show text in the interval 5-10
17098 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
17099 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
17101 # desaturate the image in the interval 15-20
17102 15.0-20.0 [enter] hue s 0,
17103 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
17105 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
17107 # apply an exponential saturation fade-out effect, starting from time 25
17108 25 [enter] hue s exp(25-t)
17111 A filtergraph allowing to read and process the above command list
17112 stored in a file @file{test.cmd}, can be specified with:
17114 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
17119 @section setpts, asetpts
17121 Change the PTS (presentation timestamp) of the input frames.
17123 @code{setpts} works on video frames, @code{asetpts} on audio frames.
17125 This filter accepts the following options:
17130 The expression which is evaluated for each frame to construct its timestamp.
17134 The expression is evaluated through the eval API and can contain the following
17139 frame rate, only defined for constant frame-rate video
17142 The presentation timestamp in input
17145 The count of the input frame for video or the number of consumed samples,
17146 not including the current frame for audio, starting from 0.
17148 @item NB_CONSUMED_SAMPLES
17149 The number of consumed samples, not including the current frame (only
17152 @item NB_SAMPLES, S
17153 The number of samples in the current frame (only audio)
17155 @item SAMPLE_RATE, SR
17156 The audio sample rate.
17159 The PTS of the first frame.
17162 the time in seconds of the first frame
17165 State whether the current frame is interlaced.
17168 the time in seconds of the current frame
17171 original position in the file of the frame, or undefined if undefined
17172 for the current frame
17175 The previous input PTS.
17178 previous input time in seconds
17181 The previous output PTS.
17184 previous output time in seconds
17187 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
17191 The wallclock (RTC) time at the start of the movie in microseconds.
17194 The timebase of the input timestamps.
17198 @subsection Examples
17202 Start counting PTS from zero
17204 setpts=PTS-STARTPTS
17208 Apply fast motion effect:
17214 Apply slow motion effect:
17220 Set fixed rate of 25 frames per second:
17226 Set fixed rate 25 fps with some jitter:
17228 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17232 Apply an offset of 10 seconds to the input PTS:
17238 Generate timestamps from a "live source" and rebase onto the current timebase:
17240 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17244 Generate timestamps by counting samples:
17251 @section settb, asettb
17253 Set the timebase to use for the output frames timestamps.
17254 It is mainly useful for testing timebase configuration.
17256 It accepts the following parameters:
17261 The expression which is evaluated into the output timebase.
17265 The value for @option{tb} is an arithmetic expression representing a
17266 rational. The expression can contain the constants "AVTB" (the default
17267 timebase), "intb" (the input timebase) and "sr" (the sample rate,
17268 audio only). Default value is "intb".
17270 @subsection Examples
17274 Set the timebase to 1/25:
17280 Set the timebase to 1/10:
17286 Set the timebase to 1001/1000:
17292 Set the timebase to 2*intb:
17298 Set the default timebase value:
17305 Convert input audio to a video output representing frequency spectrum
17306 logarithmically using Brown-Puckette constant Q transform algorithm with
17307 direct frequency domain coefficient calculation (but the transform itself
17308 is not really constant Q, instead the Q factor is actually variable/clamped),
17309 with musical tone scale, from E0 to D#10.
17311 The filter accepts the following options:
17315 Specify the video size for the output. It must be even. For the syntax of this option,
17316 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17317 Default value is @code{1920x1080}.
17320 Set the output frame rate. Default value is @code{25}.
17323 Set the bargraph height. It must be even. Default value is @code{-1} which
17324 computes the bargraph height automatically.
17327 Set the axis height. It must be even. Default value is @code{-1} which computes
17328 the axis height automatically.
17331 Set the sonogram height. It must be even. Default value is @code{-1} which
17332 computes the sonogram height automatically.
17335 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17336 instead. Default value is @code{1}.
17338 @item sono_v, volume
17339 Specify the sonogram volume expression. It can contain variables:
17342 the @var{bar_v} evaluated expression
17343 @item frequency, freq, f
17344 the frequency where it is evaluated
17345 @item timeclamp, tc
17346 the value of @var{timeclamp} option
17350 @item a_weighting(f)
17351 A-weighting of equal loudness
17352 @item b_weighting(f)
17353 B-weighting of equal loudness
17354 @item c_weighting(f)
17355 C-weighting of equal loudness.
17357 Default value is @code{16}.
17359 @item bar_v, volume2
17360 Specify the bargraph volume expression. It can contain variables:
17363 the @var{sono_v} evaluated expression
17364 @item frequency, freq, f
17365 the frequency where it is evaluated
17366 @item timeclamp, tc
17367 the value of @var{timeclamp} option
17371 @item a_weighting(f)
17372 A-weighting of equal loudness
17373 @item b_weighting(f)
17374 B-weighting of equal loudness
17375 @item c_weighting(f)
17376 C-weighting of equal loudness.
17378 Default value is @code{sono_v}.
17380 @item sono_g, gamma
17381 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17382 higher gamma makes the spectrum having more range. Default value is @code{3}.
17383 Acceptable range is @code{[1, 7]}.
17385 @item bar_g, gamma2
17386 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17390 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17391 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17393 @item timeclamp, tc
17394 Specify the transform timeclamp. At low frequency, there is trade-off between
17395 accuracy in time domain and frequency domain. If timeclamp is lower,
17396 event in time domain is represented more accurately (such as fast bass drum),
17397 otherwise event in frequency domain is represented more accurately
17398 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17401 Set attack time in seconds. The default is @code{0} (disabled). Otherwise, it
17402 limits future samples by applying asymmetric windowing in time domain, useful
17403 when low latency is required. Accepted range is @code{[0, 1]}.
17406 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17407 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17410 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17411 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17414 This option is deprecated and ignored.
17417 Specify the transform length in time domain. Use this option to control accuracy
17418 trade-off between time domain and frequency domain at every frequency sample.
17419 It can contain variables:
17421 @item frequency, freq, f
17422 the frequency where it is evaluated
17423 @item timeclamp, tc
17424 the value of @var{timeclamp} option.
17426 Default value is @code{384*tc/(384+tc*f)}.
17429 Specify the transform count for every video frame. Default value is @code{6}.
17430 Acceptable range is @code{[1, 30]}.
17433 Specify the transform count for every single pixel. Default value is @code{0},
17434 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17437 Specify font file for use with freetype to draw the axis. If not specified,
17438 use embedded font. Note that drawing with font file or embedded font is not
17439 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17443 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17444 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17447 Specify font color expression. This is arithmetic expression that should return
17448 integer value 0xRRGGBB. It can contain variables:
17450 @item frequency, freq, f
17451 the frequency where it is evaluated
17452 @item timeclamp, tc
17453 the value of @var{timeclamp} option
17458 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17459 @item r(x), g(x), b(x)
17460 red, green, and blue value of intensity x.
17462 Default value is @code{st(0, (midi(f)-59.5)/12);
17463 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17464 r(1-ld(1)) + b(ld(1))}.
17467 Specify image file to draw the axis. This option override @var{fontfile} and
17468 @var{fontcolor} option.
17471 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17472 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17473 Default value is @code{1}.
17476 Set colorspace. The accepted values are:
17479 Unspecified (default)
17488 BT.470BG or BT.601-6 625
17491 SMPTE-170M or BT.601-6 525
17497 BT.2020 with non-constant luminance
17502 Set spectrogram color scheme. This is list of floating point values with format
17503 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17504 The default is @code{1|0.5|0|0|0.5|1}.
17508 @subsection Examples
17512 Playing audio while showing the spectrum:
17514 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17518 Same as above, but with frame rate 30 fps:
17520 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17524 Playing at 1280x720:
17526 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17530 Disable sonogram display:
17536 A1 and its harmonics: A1, A2, (near)E3, A3:
17538 ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
17539 asplit[a][out1]; [a] showcqt [out0]'
17543 Same as above, but with more accuracy in frequency domain:
17545 ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
17546 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17552 bar_v=10:sono_v=bar_v*a_weighting(f)
17556 Custom gamma, now spectrum is linear to the amplitude.
17562 Custom tlength equation:
17564 tc=0.33:tlength='st(0,0.17); 384*tc / (384 / ld(0) + tc*f /(1-ld(0))) + 384*tc / (tc*f / ld(0) + 384 /(1-ld(0)))'
17568 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17570 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17574 Custom font using fontconfig:
17576 font='Courier New,Monospace,mono|bold'
17580 Custom frequency range with custom axis using image file:
17582 axisfile=myaxis.png:basefreq=40:endfreq=10000
17588 Convert input audio to video output representing the audio power spectrum.
17589 Audio amplitude is on Y-axis while frequency is on X-axis.
17591 The filter accepts the following options:
17595 Specify size of video. For the syntax of this option, check the
17596 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17597 Default is @code{1024x512}.
17601 This set how each frequency bin will be represented.
17603 It accepts the following values:
17609 Default is @code{bar}.
17612 Set amplitude scale.
17614 It accepts the following values:
17628 Default is @code{log}.
17631 Set frequency scale.
17633 It accepts the following values:
17642 Reverse logarithmic scale.
17644 Default is @code{lin}.
17649 It accepts the following values:
17665 Default is @code{w2048}
17668 Set windowing function.
17670 It accepts the following values:
17692 Default is @code{hanning}.
17695 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17696 which means optimal overlap for selected window function will be picked.
17699 Set time averaging. Setting this to 0 will display current maximal peaks.
17700 Default is @code{1}, which means time averaging is disabled.
17703 Specify list of colors separated by space or by '|' which will be used to
17704 draw channel frequencies. Unrecognized or missing colors will be replaced
17708 Set channel display mode.
17710 It accepts the following values:
17715 Default is @code{combined}.
17718 Set minimum amplitude used in @code{log} amplitude scaler.
17722 @anchor{showspectrum}
17723 @section showspectrum
17725 Convert input audio to a video output, representing the audio frequency
17728 The filter accepts the following options:
17732 Specify the video size for the output. For the syntax of this option, check the
17733 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17734 Default value is @code{640x512}.
17737 Specify how the spectrum should slide along the window.
17739 It accepts the following values:
17742 the samples start again on the left when they reach the right
17744 the samples scroll from right to left
17746 frames are only produced when the samples reach the right
17748 the samples scroll from left to right
17751 Default value is @code{replace}.
17754 Specify display mode.
17756 It accepts the following values:
17759 all channels are displayed in the same row
17761 all channels are displayed in separate rows
17764 Default value is @samp{combined}.
17767 Specify display color mode.
17769 It accepts the following values:
17772 each channel is displayed in a separate color
17774 each channel is displayed using the same color scheme
17776 each channel is displayed using the rainbow color scheme
17778 each channel is displayed using the moreland color scheme
17780 each channel is displayed using the nebulae color scheme
17782 each channel is displayed using the fire color scheme
17784 each channel is displayed using the fiery color scheme
17786 each channel is displayed using the fruit color scheme
17788 each channel is displayed using the cool color scheme
17791 Default value is @samp{channel}.
17794 Specify scale used for calculating intensity color values.
17796 It accepts the following values:
17801 square root, default
17812 Default value is @samp{sqrt}.
17815 Set saturation modifier for displayed colors. Negative values provide
17816 alternative color scheme. @code{0} is no saturation at all.
17817 Saturation must be in [-10.0, 10.0] range.
17818 Default value is @code{1}.
17821 Set window function.
17823 It accepts the following values:
17847 Default value is @code{hann}.
17850 Set orientation of time vs frequency axis. Can be @code{vertical} or
17851 @code{horizontal}. Default is @code{vertical}.
17854 Set ratio of overlap window. Default value is @code{0}.
17855 When value is @code{1} overlap is set to recommended size for specific
17856 window function currently used.
17859 Set scale gain for calculating intensity color values.
17860 Default value is @code{1}.
17863 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17866 Set color rotation, must be in [-1.0, 1.0] range.
17867 Default value is @code{0}.
17870 The usage is very similar to the showwaves filter; see the examples in that
17873 @subsection Examples
17877 Large window with logarithmic color scaling:
17879 showspectrum=s=1280x480:scale=log
17883 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17885 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17886 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17890 @section showspectrumpic
17892 Convert input audio to a single video frame, representing the audio frequency
17895 The filter accepts the following options:
17899 Specify the video size for the output. For the syntax of this option, check the
17900 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17901 Default value is @code{4096x2048}.
17904 Specify display mode.
17906 It accepts the following values:
17909 all channels are displayed in the same row
17911 all channels are displayed in separate rows
17913 Default value is @samp{combined}.
17916 Specify display color mode.
17918 It accepts the following values:
17921 each channel is displayed in a separate color
17923 each channel is displayed using the same color scheme
17925 each channel is displayed using the rainbow color scheme
17927 each channel is displayed using the moreland color scheme
17929 each channel is displayed using the nebulae color scheme
17931 each channel is displayed using the fire color scheme
17933 each channel is displayed using the fiery color scheme
17935 each channel is displayed using the fruit color scheme
17937 each channel is displayed using the cool color scheme
17939 Default value is @samp{intensity}.
17942 Specify scale used for calculating intensity color values.
17944 It accepts the following values:
17949 square root, default
17959 Default value is @samp{log}.
17962 Set saturation modifier for displayed colors. Negative values provide
17963 alternative color scheme. @code{0} is no saturation at all.
17964 Saturation must be in [-10.0, 10.0] range.
17965 Default value is @code{1}.
17968 Set window function.
17970 It accepts the following values:
17993 Default value is @code{hann}.
17996 Set orientation of time vs frequency axis. Can be @code{vertical} or
17997 @code{horizontal}. Default is @code{vertical}.
18000 Set scale gain for calculating intensity color values.
18001 Default value is @code{1}.
18004 Draw time and frequency axes and legends. Default is enabled.
18007 Set color rotation, must be in [-1.0, 1.0] range.
18008 Default value is @code{0}.
18011 @subsection Examples
18015 Extract an audio spectrogram of a whole audio track
18016 in a 1024x1024 picture using @command{ffmpeg}:
18018 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
18022 @section showvolume
18024 Convert input audio volume to a video output.
18026 The filter accepts the following options:
18033 Set border width, allowed range is [0, 5]. Default is 1.
18036 Set channel width, allowed range is [80, 8192]. Default is 400.
18039 Set channel height, allowed range is [1, 900]. Default is 20.
18042 Set fade, allowed range is [0.001, 1]. Default is 0.95.
18045 Set volume color expression.
18047 The expression can use the following variables:
18051 Current max volume of channel in dB.
18057 Current channel number, starting from 0.
18061 If set, displays channel names. Default is enabled.
18064 If set, displays volume values. Default is enabled.
18067 Set orientation, can be @code{horizontal} or @code{vertical},
18068 default is @code{horizontal}.
18071 Set step size, allowed range s [0, 5]. Default is 0, which means
18077 Convert input audio to a video output, representing the samples waves.
18079 The filter accepts the following options:
18083 Specify the video size for the output. For the syntax of this option, check the
18084 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
18085 Default value is @code{600x240}.
18090 Available values are:
18093 Draw a point for each sample.
18096 Draw a vertical line for each sample.
18099 Draw a point for each sample and a line between them.
18102 Draw a centered vertical line for each sample.
18105 Default value is @code{point}.
18108 Set the number of samples which are printed on the same column. A
18109 larger value will decrease the frame rate. Must be a positive
18110 integer. This option can be set only if the value for @var{rate}
18111 is not explicitly specified.
18114 Set the (approximate) output frame rate. This is done by setting the
18115 option @var{n}. Default value is "25".
18117 @item split_channels
18118 Set if channels should be drawn separately or overlap. Default value is 0.
18121 Set colors separated by '|' which are going to be used for drawing of each channel.
18124 Set amplitude scale.
18126 Available values are:
18144 @subsection Examples
18148 Output the input file audio and the corresponding video representation
18151 amovie=a.mp3,asplit[out0],showwaves[out1]
18155 Create a synthetic signal and show it with showwaves, forcing a
18156 frame rate of 30 frames per second:
18158 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
18162 @section showwavespic
18164 Convert input audio to a single video frame, representing the samples waves.
18166 The filter accepts the following options:
18170 Specify the video size for the output. For the syntax of this option, check the
18171 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
18172 Default value is @code{600x240}.
18174 @item split_channels
18175 Set if channels should be drawn separately or overlap. Default value is 0.
18178 Set colors separated by '|' which are going to be used for drawing of each channel.
18181 Set amplitude scale.
18183 Available values are:
18201 @subsection Examples
18205 Extract a channel split representation of the wave form of a whole audio track
18206 in a 1024x800 picture using @command{ffmpeg}:
18208 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
18212 @section sidedata, asidedata
18214 Delete frame side data, or select frames based on it.
18216 This filter accepts the following options:
18220 Set mode of operation of the filter.
18222 Can be one of the following:
18226 Select every frame with side data of @code{type}.
18229 Delete side data of @code{type}. If @code{type} is not set, delete all side
18235 Set side data type used with all modes. Must be set for @code{select} mode. For
18236 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18237 in @file{libavutil/frame.h}. For example, to choose
18238 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18242 @section spectrumsynth
18244 Sythesize audio from 2 input video spectrums, first input stream represents
18245 magnitude across time and second represents phase across time.
18246 The filter will transform from frequency domain as displayed in videos back
18247 to time domain as presented in audio output.
18249 This filter is primarily created for reversing processed @ref{showspectrum}
18250 filter outputs, but can synthesize sound from other spectrograms too.
18251 But in such case results are going to be poor if the phase data is not
18252 available, because in such cases phase data need to be recreated, usually
18253 its just recreated from random noise.
18254 For best results use gray only output (@code{channel} color mode in
18255 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
18256 @code{lin} scale for phase video. To produce phase, for 2nd video, use
18257 @code{data} option. Inputs videos should generally use @code{fullframe}
18258 slide mode as that saves resources needed for decoding video.
18260 The filter accepts the following options:
18264 Specify sample rate of output audio, the sample rate of audio from which
18265 spectrum was generated may differ.
18268 Set number of channels represented in input video spectrums.
18271 Set scale which was used when generating magnitude input spectrum.
18272 Can be @code{lin} or @code{log}. Default is @code{log}.
18275 Set slide which was used when generating inputs spectrums.
18276 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18277 Default is @code{fullframe}.
18280 Set window function used for resynthesis.
18283 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18284 which means optimal overlap for selected window function will be picked.
18287 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18288 Default is @code{vertical}.
18291 @subsection Examples
18295 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18296 then resynthesize videos back to audio with spectrumsynth:
18298 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=log:overlap=0.875:color=channel:slide=fullframe:data=magnitude -an -c:v rawvideo magnitude.nut
18299 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=lin:overlap=0.875:color=channel:slide=fullframe:data=phase -an -c:v rawvideo phase.nut
18300 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18304 @section split, asplit
18306 Split input into several identical outputs.
18308 @code{asplit} works with audio input, @code{split} with video.
18310 The filter accepts a single parameter which specifies the number of outputs. If
18311 unspecified, it defaults to 2.
18313 @subsection Examples
18317 Create two separate outputs from the same input:
18319 [in] split [out0][out1]
18323 To create 3 or more outputs, you need to specify the number of
18326 [in] asplit=3 [out0][out1][out2]
18330 Create two separate outputs from the same input, one cropped and
18333 [in] split [splitout1][splitout2];
18334 [splitout1] crop=100:100:0:0 [cropout];
18335 [splitout2] pad=200:200:100:100 [padout];
18339 Create 5 copies of the input audio with @command{ffmpeg}:
18341 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18347 Receive commands sent through a libzmq client, and forward them to
18348 filters in the filtergraph.
18350 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18351 must be inserted between two video filters, @code{azmq} between two
18354 To enable these filters you need to install the libzmq library and
18355 headers and configure FFmpeg with @code{--enable-libzmq}.
18357 For more information about libzmq see:
18358 @url{http://www.zeromq.org/}
18360 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18361 receives messages sent through a network interface defined by the
18362 @option{bind_address} option.
18364 The received message must be in the form:
18366 @var{TARGET} @var{COMMAND} [@var{ARG}]
18369 @var{TARGET} specifies the target of the command, usually the name of
18370 the filter class or a specific filter instance name.
18372 @var{COMMAND} specifies the name of the command for the target filter.
18374 @var{ARG} is optional and specifies the optional argument list for the
18375 given @var{COMMAND}.
18377 Upon reception, the message is processed and the corresponding command
18378 is injected into the filtergraph. Depending on the result, the filter
18379 will send a reply to the client, adopting the format:
18381 @var{ERROR_CODE} @var{ERROR_REASON}
18385 @var{MESSAGE} is optional.
18387 @subsection Examples
18389 Look at @file{tools/zmqsend} for an example of a zmq client which can
18390 be used to send commands processed by these filters.
18392 Consider the following filtergraph generated by @command{ffplay}
18394 ffplay -dumpgraph 1 -f lavfi "
18395 color=s=100x100:c=red [l];
18396 color=s=100x100:c=blue [r];
18397 nullsrc=s=200x100, zmq [bg];
18398 [bg][l] overlay [bg+l];
18399 [bg+l][r] overlay=x=100 "
18402 To change the color of the left side of the video, the following
18403 command can be used:
18405 echo Parsed_color_0 c yellow | tools/zmqsend
18408 To change the right side:
18410 echo Parsed_color_1 c pink | tools/zmqsend
18413 @c man end MULTIMEDIA FILTERS
18415 @chapter Multimedia Sources
18416 @c man begin MULTIMEDIA SOURCES
18418 Below is a description of the currently available multimedia sources.
18422 This is the same as @ref{movie} source, except it selects an audio
18428 Read audio and/or video stream(s) from a movie container.
18430 It accepts the following parameters:
18434 The name of the resource to read (not necessarily a file; it can also be a
18435 device or a stream accessed through some protocol).
18437 @item format_name, f
18438 Specifies the format assumed for the movie to read, and can be either
18439 the name of a container or an input device. If not specified, the
18440 format is guessed from @var{movie_name} or by probing.
18442 @item seek_point, sp
18443 Specifies the seek point in seconds. The frames will be output
18444 starting from this seek point. The parameter is evaluated with
18445 @code{av_strtod}, so the numerical value may be suffixed by an IS
18446 postfix. The default value is "0".
18449 Specifies the streams to read. Several streams can be specified,
18450 separated by "+". The source will then have as many outputs, in the
18451 same order. The syntax is explained in the ``Stream specifiers''
18452 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18453 respectively the default (best suited) video and audio stream. Default
18454 is "dv", or "da" if the filter is called as "amovie".
18456 @item stream_index, si
18457 Specifies the index of the video stream to read. If the value is -1,
18458 the most suitable video stream will be automatically selected. The default
18459 value is "-1". Deprecated. If the filter is called "amovie", it will select
18460 audio instead of video.
18463 Specifies how many times to read the stream in sequence.
18464 If the value is 0, the stream will be looped infinitely.
18465 Default value is "1".
18467 Note that when the movie is looped the source timestamps are not
18468 changed, so it will generate non monotonically increasing timestamps.
18470 @item discontinuity
18471 Specifies the time difference between frames above which the point is
18472 considered a timestamp discontinuity which is removed by adjusting the later
18476 It allows overlaying a second video on top of the main input of
18477 a filtergraph, as shown in this graph:
18479 input -----------> deltapts0 --> overlay --> output
18482 movie --> scale--> deltapts1 -------+
18484 @subsection Examples
18488 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18489 on top of the input labelled "in":
18491 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18492 [in] setpts=PTS-STARTPTS [main];
18493 [main][over] overlay=16:16 [out]
18497 Read from a video4linux2 device, and overlay it on top of the input
18500 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18501 [in] setpts=PTS-STARTPTS [main];
18502 [main][over] overlay=16:16 [out]
18506 Read the first video stream and the audio stream with id 0x81 from
18507 dvd.vob; the video is connected to the pad named "video" and the audio is
18508 connected to the pad named "audio":
18510 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18514 @subsection Commands
18516 Both movie and amovie support the following commands:
18519 Perform seek using "av_seek_frame".
18520 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18523 @var{stream_index}: If stream_index is -1, a default
18524 stream is selected, and @var{timestamp} is automatically converted
18525 from AV_TIME_BASE units to the stream specific time_base.
18527 @var{timestamp}: Timestamp in AVStream.time_base units
18528 or, if no stream is specified, in AV_TIME_BASE units.
18530 @var{flags}: Flags which select direction and seeking mode.
18534 Get movie duration in AV_TIME_BASE units.
18538 @c man end MULTIMEDIA SOURCES