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.
3008 Mix channels with specific gain levels. The filter accepts the output
3009 channel layout followed by a set of channels definitions.
3011 This filter is also designed to efficiently remap the channels of an audio
3014 The filter accepts parameters of the form:
3015 "@var{l}|@var{outdef}|@var{outdef}|..."
3019 output channel layout or number of channels
3022 output channel specification, of the form:
3023 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3026 output channel to define, either a channel name (FL, FR, etc.) or a channel
3027 number (c0, c1, etc.)
3030 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3033 input channel to use, see out_name for details; it is not possible to mix
3034 named and numbered input channels
3037 If the `=' in a channel specification is replaced by `<', then the gains for
3038 that specification will be renormalized so that the total is 1, thus
3039 avoiding clipping noise.
3041 @subsection Mixing examples
3043 For example, if you want to down-mix from stereo to mono, but with a bigger
3044 factor for the left channel:
3046 pan=1c|c0=0.9*c0+0.1*c1
3049 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3050 7-channels surround:
3052 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3055 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3056 that should be preferred (see "-ac" option) unless you have very specific
3059 @subsection Remapping examples
3061 The channel remapping will be effective if, and only if:
3064 @item gain coefficients are zeroes or ones,
3065 @item only one input per channel output,
3068 If all these conditions are satisfied, the filter will notify the user ("Pure
3069 channel mapping detected"), and use an optimized and lossless method to do the
3072 For example, if you have a 5.1 source and want a stereo audio stream by
3073 dropping the extra channels:
3075 pan="stereo| c0=FL | c1=FR"
3078 Given the same source, you can also switch front left and front right channels
3079 and keep the input channel layout:
3081 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3084 If the input is a stereo audio stream, you can mute the front left channel (and
3085 still keep the stereo channel layout) with:
3090 Still with a stereo audio stream input, you can copy the right channel in both
3091 front left and right:
3093 pan="stereo| c0=FR | c1=FR"
3098 ReplayGain scanner filter. This filter takes an audio stream as an input and
3099 outputs it unchanged.
3100 At end of filtering it displays @code{track_gain} and @code{track_peak}.
3104 Convert the audio sample format, sample rate and channel layout. It is
3105 not meant to be used directly.
3108 Apply time-stretching and pitch-shifting with librubberband.
3110 The filter accepts the following options:
3114 Set tempo scale factor.
3117 Set pitch scale factor.
3120 Set transients detector.
3121 Possible values are:
3130 Possible values are:
3139 Possible values are:
3146 Set processing window size.
3147 Possible values are:
3156 Possible values are:
3163 Enable formant preservation when shift pitching.
3164 Possible values are:
3172 Possible values are:
3181 Possible values are:
3188 @section sidechaincompress
3190 This filter acts like normal compressor but has the ability to compress
3191 detected signal using second input signal.
3192 It needs two input streams and returns one output stream.
3193 First input stream will be processed depending on second stream signal.
3194 The filtered signal then can be filtered with other filters in later stages of
3195 processing. See @ref{pan} and @ref{amerge} filter.
3197 The filter accepts the following options:
3201 Set input gain. Default is 1. Range is between 0.015625 and 64.
3204 If a signal of second stream raises above this level it will affect the gain
3205 reduction of first stream.
3206 By default is 0.125. Range is between 0.00097563 and 1.
3209 Set a ratio about which the signal is reduced. 1:2 means that if the level
3210 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3211 Default is 2. Range is between 1 and 20.
3214 Amount of milliseconds the signal has to rise above the threshold before gain
3215 reduction starts. Default is 20. Range is between 0.01 and 2000.
3218 Amount of milliseconds the signal has to fall below the threshold before
3219 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3222 Set the amount by how much signal will be amplified after processing.
3223 Default is 2. Range is from 1 and 64.
3226 Curve the sharp knee around the threshold to enter gain reduction more softly.
3227 Default is 2.82843. Range is between 1 and 8.
3230 Choose if the @code{average} level between all channels of side-chain stream
3231 or the louder(@code{maximum}) channel of side-chain stream affects the
3232 reduction. Default is @code{average}.
3235 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3236 of @code{rms}. Default is @code{rms} which is mainly smoother.
3239 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3242 How much to use compressed signal in output. Default is 1.
3243 Range is between 0 and 1.
3246 @subsection Examples
3250 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3251 depending on the signal of 2nd input and later compressed signal to be
3252 merged with 2nd input:
3254 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3258 @section sidechaingate
3260 A sidechain gate acts like a normal (wideband) gate but has the ability to
3261 filter the detected signal before sending it to the gain reduction stage.
3262 Normally a gate uses the full range signal to detect a level above the
3264 For example: If you cut all lower frequencies from your sidechain signal
3265 the gate will decrease the volume of your track only if not enough highs
3266 appear. With this technique you are able to reduce the resonation of a
3267 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3269 It needs two input streams and returns one output stream.
3270 First input stream will be processed depending on second stream signal.
3272 The filter accepts the following options:
3276 Set input level before filtering.
3277 Default is 1. Allowed range is from 0.015625 to 64.
3280 Set the level of gain reduction when the signal is below the threshold.
3281 Default is 0.06125. Allowed range is from 0 to 1.
3284 If a signal rises above this level the gain reduction is released.
3285 Default is 0.125. Allowed range is from 0 to 1.
3288 Set a ratio about which the signal is reduced.
3289 Default is 2. Allowed range is from 1 to 9000.
3292 Amount of milliseconds the signal has to rise above the threshold before gain
3294 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3297 Amount of milliseconds the signal has to fall below the threshold before the
3298 reduction is increased again. Default is 250 milliseconds.
3299 Allowed range is from 0.01 to 9000.
3302 Set amount of amplification of signal after processing.
3303 Default is 1. Allowed range is from 1 to 64.
3306 Curve the sharp knee around the threshold to enter gain reduction more softly.
3307 Default is 2.828427125. Allowed range is from 1 to 8.
3310 Choose if exact signal should be taken for detection or an RMS like one.
3311 Default is rms. Can be peak or rms.
3314 Choose if the average level between all channels or the louder channel affects
3316 Default is average. Can be average or maximum.
3319 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3322 @section silencedetect
3324 Detect silence in an audio stream.
3326 This filter logs a message when it detects that the input audio volume is less
3327 or equal to a noise tolerance value for a duration greater or equal to the
3328 minimum detected noise duration.
3330 The printed times and duration are expressed in seconds.
3332 The filter accepts the following options:
3336 Set silence duration until notification (default is 2 seconds).
3339 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3340 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3343 @subsection Examples
3347 Detect 5 seconds of silence with -50dB noise tolerance:
3349 silencedetect=n=-50dB:d=5
3353 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3354 tolerance in @file{silence.mp3}:
3356 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3360 @section silenceremove
3362 Remove silence from the beginning, middle or end of the audio.
3364 The filter accepts the following options:
3368 This value is used to indicate if audio should be trimmed at beginning of
3369 the audio. A value of zero indicates no silence should be trimmed from the
3370 beginning. When specifying a non-zero value, it trims audio up until it
3371 finds non-silence. Normally, when trimming silence from beginning of audio
3372 the @var{start_periods} will be @code{1} but it can be increased to higher
3373 values to trim all audio up to specific count of non-silence periods.
3374 Default value is @code{0}.
3376 @item start_duration
3377 Specify the amount of time that non-silence must be detected before it stops
3378 trimming audio. By increasing the duration, bursts of noises can be treated
3379 as silence and trimmed off. Default value is @code{0}.
3381 @item start_threshold
3382 This indicates what sample value should be treated as silence. For digital
3383 audio, a value of @code{0} may be fine but for audio recorded from analog,
3384 you may wish to increase the value to account for background noise.
3385 Can be specified in dB (in case "dB" is appended to the specified value)
3386 or amplitude ratio. Default value is @code{0}.
3389 Set the count for trimming silence from the end of audio.
3390 To remove silence from the middle of a file, specify a @var{stop_periods}
3391 that is negative. This value is then treated as a positive value and is
3392 used to indicate the effect should restart processing as specified by
3393 @var{start_periods}, making it suitable for removing periods of silence
3394 in the middle of the audio.
3395 Default value is @code{0}.
3398 Specify a duration of silence that must exist before audio is not copied any
3399 more. By specifying a higher duration, silence that is wanted can be left in
3401 Default value is @code{0}.
3403 @item stop_threshold
3404 This is the same as @option{start_threshold} but for trimming silence from
3406 Can be specified in dB (in case "dB" is appended to the specified value)
3407 or amplitude ratio. Default value is @code{0}.
3410 This indicates that @var{stop_duration} length of audio should be left intact
3411 at the beginning of each period of silence.
3412 For example, if you want to remove long pauses between words but do not want
3413 to remove the pauses completely. Default value is @code{0}.
3416 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3417 and works better with digital silence which is exactly 0.
3418 Default value is @code{rms}.
3421 Set ratio used to calculate size of window for detecting silence.
3422 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3425 @subsection Examples
3429 The following example shows how this filter can be used to start a recording
3430 that does not contain the delay at the start which usually occurs between
3431 pressing the record button and the start of the performance:
3433 silenceremove=1:5:0.02
3437 Trim all silence encountered from beginning to end where there is more than 1
3438 second of silence in audio:
3440 silenceremove=0:0:0:-1:1:-90dB
3446 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3447 loudspeakers around the user for binaural listening via headphones (audio
3448 formats up to 9 channels supported).
3449 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3450 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3451 Austrian Academy of Sciences.
3453 To enable compilation of this filter you need to configure FFmpeg with
3454 @code{--enable-netcdf}.
3456 The filter accepts the following options:
3460 Set the SOFA file used for rendering.
3463 Set gain applied to audio. Value is in dB. Default is 0.
3466 Set rotation of virtual loudspeakers in deg. Default is 0.
3469 Set elevation of virtual speakers in deg. Default is 0.
3472 Set distance in meters between loudspeakers and the listener with near-field
3473 HRTFs. Default is 1.
3476 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3477 processing audio in time domain which is slow.
3478 @var{freq} is processing audio in frequency domain which is fast.
3479 Default is @var{freq}.
3482 Set custom positions of virtual loudspeakers. Syntax for this option is:
3483 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3484 Each virtual loudspeaker is described with short channel name following with
3485 azimuth and elevation in degreees.
3486 Each virtual loudspeaker description is separated by '|'.
3487 For example to override front left and front right channel positions use:
3488 'speakers=FL 45 15|FR 345 15'.
3489 Descriptions with unrecognised channel names are ignored.
3492 @subsection Examples
3496 Using ClubFritz6 sofa file:
3498 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3502 Using ClubFritz12 sofa file and bigger radius with small rotation:
3504 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3508 Similar as above but with custom speaker positions for front left, front right, back left and back right
3509 and also with custom gain:
3511 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3515 @section stereotools
3517 This filter has some handy utilities to manage stereo signals, for converting
3518 M/S stereo recordings to L/R signal while having control over the parameters
3519 or spreading the stereo image of master track.
3521 The filter accepts the following options:
3525 Set input level before filtering for both channels. Defaults is 1.
3526 Allowed range is from 0.015625 to 64.
3529 Set output level after filtering for both channels. Defaults is 1.
3530 Allowed range is from 0.015625 to 64.
3533 Set input balance between both channels. Default is 0.
3534 Allowed range is from -1 to 1.
3537 Set output balance between both channels. Default is 0.
3538 Allowed range is from -1 to 1.
3541 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3542 clipping. Disabled by default.
3545 Mute the left channel. Disabled by default.
3548 Mute the right channel. Disabled by default.
3551 Change the phase of the left channel. Disabled by default.
3554 Change the phase of the right channel. Disabled by default.
3557 Set stereo mode. Available values are:
3561 Left/Right to Left/Right, this is default.
3564 Left/Right to Mid/Side.
3567 Mid/Side to Left/Right.
3570 Left/Right to Left/Left.
3573 Left/Right to Right/Right.
3576 Left/Right to Left + Right.
3579 Left/Right to Right/Left.
3583 Set level of side signal. Default is 1.
3584 Allowed range is from 0.015625 to 64.
3587 Set balance of side signal. Default is 0.
3588 Allowed range is from -1 to 1.
3591 Set level of the middle signal. Default is 1.
3592 Allowed range is from 0.015625 to 64.
3595 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3598 Set stereo base between mono and inversed channels. Default is 0.
3599 Allowed range is from -1 to 1.
3602 Set delay in milliseconds how much to delay left from right channel and
3603 vice versa. Default is 0. Allowed range is from -20 to 20.
3606 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3609 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3612 @subsection Examples
3616 Apply karaoke like effect:
3618 stereotools=mlev=0.015625
3622 Convert M/S signal to L/R:
3624 "stereotools=mode=ms>lr"
3628 @section stereowiden
3630 This filter enhance the stereo effect by suppressing signal common to both
3631 channels and by delaying the signal of left into right and vice versa,
3632 thereby widening the stereo effect.
3634 The filter accepts the following options:
3638 Time in milliseconds of the delay of left signal into right and vice versa.
3639 Default is 20 milliseconds.
3642 Amount of gain in delayed signal into right and vice versa. Gives a delay
3643 effect of left signal in right output and vice versa which gives widening
3644 effect. Default is 0.3.
3647 Cross feed of left into right with inverted phase. This helps in suppressing
3648 the mono. If the value is 1 it will cancel all the signal common to both
3649 channels. Default is 0.3.
3652 Set level of input signal of original channel. Default is 0.8.
3657 Boost or cut treble (upper) frequencies of the audio using a two-pole
3658 shelving filter with a response similar to that of a standard
3659 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3661 The filter accepts the following options:
3665 Give the gain at whichever is the lower of ~22 kHz and the
3666 Nyquist frequency. Its useful range is about -20 (for a large cut)
3667 to +20 (for a large boost). Beware of clipping when using a positive gain.
3670 Set the filter's central frequency and so can be used
3671 to extend or reduce the frequency range to be boosted or cut.
3672 The default value is @code{3000} Hz.
3675 Set method to specify band-width of filter.
3688 Determine how steep is the filter's shelf transition.
3691 Specify which channels to filter, by default all available are filtered.
3696 Sinusoidal amplitude modulation.
3698 The filter accepts the following options:
3702 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3703 (20 Hz or lower) will result in a tremolo effect.
3704 This filter may also be used as a ring modulator by specifying
3705 a modulation frequency higher than 20 Hz.
3706 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3709 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3710 Default value is 0.5.
3715 Sinusoidal phase modulation.
3717 The filter accepts the following options:
3721 Modulation frequency in Hertz.
3722 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3725 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3726 Default value is 0.5.
3731 Adjust the input audio volume.
3733 It accepts the following parameters:
3737 Set audio volume expression.
3739 Output values are clipped to the maximum value.
3741 The output audio volume is given by the relation:
3743 @var{output_volume} = @var{volume} * @var{input_volume}
3746 The default value for @var{volume} is "1.0".
3749 This parameter represents the mathematical precision.
3751 It determines which input sample formats will be allowed, which affects the
3752 precision of the volume scaling.
3756 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3758 32-bit floating-point; this limits input sample format to FLT. (default)
3760 64-bit floating-point; this limits input sample format to DBL.
3764 Choose the behaviour on encountering ReplayGain side data in input frames.
3768 Remove ReplayGain side data, ignoring its contents (the default).
3771 Ignore ReplayGain side data, but leave it in the frame.
3774 Prefer the track gain, if present.
3777 Prefer the album gain, if present.
3780 @item replaygain_preamp
3781 Pre-amplification gain in dB to apply to the selected replaygain gain.
3783 Default value for @var{replaygain_preamp} is 0.0.
3786 Set when the volume expression is evaluated.
3788 It accepts the following values:
3791 only evaluate expression once during the filter initialization, or
3792 when the @samp{volume} command is sent
3795 evaluate expression for each incoming frame
3798 Default value is @samp{once}.
3801 The volume expression can contain the following parameters.
3805 frame number (starting at zero)
3808 @item nb_consumed_samples
3809 number of samples consumed by the filter
3811 number of samples in the current frame
3813 original frame position in the file
3819 PTS at start of stream
3821 time at start of stream
3827 last set volume value
3830 Note that when @option{eval} is set to @samp{once} only the
3831 @var{sample_rate} and @var{tb} variables are available, all other
3832 variables will evaluate to NAN.
3834 @subsection Commands
3836 This filter supports the following commands:
3839 Modify the volume expression.
3840 The command accepts the same syntax of the corresponding option.
3842 If the specified expression is not valid, it is kept at its current
3844 @item replaygain_noclip
3845 Prevent clipping by limiting the gain applied.
3847 Default value for @var{replaygain_noclip} is 1.
3851 @subsection Examples
3855 Halve the input audio volume:
3859 volume=volume=-6.0206dB
3862 In all the above example the named key for @option{volume} can be
3863 omitted, for example like in:
3869 Increase input audio power by 6 decibels using fixed-point precision:
3871 volume=volume=6dB:precision=fixed
3875 Fade volume after time 10 with an annihilation period of 5 seconds:
3877 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3881 @section volumedetect
3883 Detect the volume of the input video.
3885 The filter has no parameters. The input is not modified. Statistics about
3886 the volume will be printed in the log when the input stream end is reached.
3888 In particular it will show the mean volume (root mean square), maximum
3889 volume (on a per-sample basis), and the beginning of a histogram of the
3890 registered volume values (from the maximum value to a cumulated 1/1000 of
3893 All volumes are in decibels relative to the maximum PCM value.
3895 @subsection Examples
3897 Here is an excerpt of the output:
3899 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3900 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3901 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3902 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3903 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3904 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3905 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3906 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3907 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3913 The mean square energy is approximately -27 dB, or 10^-2.7.
3915 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3917 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3920 In other words, raising the volume by +4 dB does not cause any clipping,
3921 raising it by +5 dB causes clipping for 6 samples, etc.
3923 @c man end AUDIO FILTERS
3925 @chapter Audio Sources
3926 @c man begin AUDIO SOURCES
3928 Below is a description of the currently available audio sources.
3932 Buffer audio frames, and make them available to the filter chain.
3934 This source is mainly intended for a programmatic use, in particular
3935 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3937 It accepts the following parameters:
3941 The timebase which will be used for timestamps of submitted frames. It must be
3942 either a floating-point number or in @var{numerator}/@var{denominator} form.
3945 The sample rate of the incoming audio buffers.
3948 The sample format of the incoming audio buffers.
3949 Either a sample format name or its corresponding integer representation from
3950 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3952 @item channel_layout
3953 The channel layout of the incoming audio buffers.
3954 Either a channel layout name from channel_layout_map in
3955 @file{libavutil/channel_layout.c} or its corresponding integer representation
3956 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3959 The number of channels of the incoming audio buffers.
3960 If both @var{channels} and @var{channel_layout} are specified, then they
3965 @subsection Examples
3968 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3971 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3972 Since the sample format with name "s16p" corresponds to the number
3973 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3976 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3981 Generate an audio signal specified by an expression.
3983 This source accepts in input one or more expressions (one for each
3984 channel), which are evaluated and used to generate a corresponding
3987 This source accepts the following options:
3991 Set the '|'-separated expressions list for each separate channel. In case the
3992 @option{channel_layout} option is not specified, the selected channel layout
3993 depends on the number of provided expressions. Otherwise the last
3994 specified expression is applied to the remaining output channels.
3996 @item channel_layout, c
3997 Set the channel layout. The number of channels in the specified layout
3998 must be equal to the number of specified expressions.
4001 Set the minimum duration of the sourced audio. See
4002 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
4003 for the accepted syntax.
4004 Note that the resulting duration may be greater than the specified
4005 duration, as the generated audio is always cut at the end of a
4008 If not specified, or the expressed duration is negative, the audio is
4009 supposed to be generated forever.
4012 Set the number of samples per channel per each output frame,
4015 @item sample_rate, s
4016 Specify the sample rate, default to 44100.
4019 Each expression in @var{exprs} can contain the following constants:
4023 number of the evaluated sample, starting from 0
4026 time of the evaluated sample expressed in seconds, starting from 0
4033 @subsection Examples
4043 Generate a sin signal with frequency of 440 Hz, set sample rate to
4046 aevalsrc="sin(440*2*PI*t):s=8000"
4050 Generate a two channels signal, specify the channel layout (Front
4051 Center + Back Center) explicitly:
4053 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4057 Generate white noise:
4059 aevalsrc="-2+random(0)"
4063 Generate an amplitude modulated signal:
4065 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4069 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4071 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4078 The null audio source, return unprocessed audio frames. It is mainly useful
4079 as a template and to be employed in analysis / debugging tools, or as
4080 the source for filters which ignore the input data (for example the sox
4083 This source accepts the following options:
4087 @item channel_layout, cl
4089 Specifies the channel layout, and can be either an integer or a string
4090 representing a channel layout. The default value of @var{channel_layout}
4093 Check the channel_layout_map definition in
4094 @file{libavutil/channel_layout.c} for the mapping between strings and
4095 channel layout values.
4097 @item sample_rate, r
4098 Specifies the sample rate, and defaults to 44100.
4101 Set the number of samples per requested frames.
4105 @subsection Examples
4109 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4111 anullsrc=r=48000:cl=4
4115 Do the same operation with a more obvious syntax:
4117 anullsrc=r=48000:cl=mono
4121 All the parameters need to be explicitly defined.
4125 Synthesize a voice utterance using the libflite library.
4127 To enable compilation of this filter you need to configure FFmpeg with
4128 @code{--enable-libflite}.
4130 Note that the flite library is not thread-safe.
4132 The filter accepts the following options:
4137 If set to 1, list the names of the available voices and exit
4138 immediately. Default value is 0.
4141 Set the maximum number of samples per frame. Default value is 512.
4144 Set the filename containing the text to speak.
4147 Set the text to speak.
4150 Set the voice to use for the speech synthesis. Default value is
4151 @code{kal}. See also the @var{list_voices} option.
4154 @subsection Examples
4158 Read from file @file{speech.txt}, and synthesize the text using the
4159 standard flite voice:
4161 flite=textfile=speech.txt
4165 Read the specified text selecting the @code{slt} voice:
4167 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4171 Input text to ffmpeg:
4173 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4177 Make @file{ffplay} speak the specified text, using @code{flite} and
4178 the @code{lavfi} device:
4180 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4184 For more information about libflite, check:
4185 @url{http://www.speech.cs.cmu.edu/flite/}
4189 Generate a noise audio signal.
4191 The filter accepts the following options:
4194 @item sample_rate, r
4195 Specify the sample rate. Default value is 48000 Hz.
4198 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4202 Specify the duration of the generated audio stream. Not specifying this option
4203 results in noise with an infinite length.
4205 @item color, colour, c
4206 Specify the color of noise. Available noise colors are white, pink, and brown.
4207 Default color is white.
4210 Specify a value used to seed the PRNG.
4213 Set the number of samples per each output frame, default is 1024.
4216 @subsection Examples
4221 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4223 anoisesrc=d=60:c=pink:r=44100:a=0.5
4229 Generate an audio signal made of a sine wave with amplitude 1/8.
4231 The audio signal is bit-exact.
4233 The filter accepts the following options:
4238 Set the carrier frequency. Default is 440 Hz.
4240 @item beep_factor, b
4241 Enable a periodic beep every second with frequency @var{beep_factor} times
4242 the carrier frequency. Default is 0, meaning the beep is disabled.
4244 @item sample_rate, r
4245 Specify the sample rate, default is 44100.
4248 Specify the duration of the generated audio stream.
4250 @item samples_per_frame
4251 Set the number of samples per output frame.
4253 The expression can contain the following constants:
4257 The (sequential) number of the output audio frame, starting from 0.
4260 The PTS (Presentation TimeStamp) of the output audio frame,
4261 expressed in @var{TB} units.
4264 The PTS of the output audio frame, expressed in seconds.
4267 The timebase of the output audio frames.
4270 Default is @code{1024}.
4273 @subsection Examples
4278 Generate a simple 440 Hz sine wave:
4284 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4288 sine=frequency=220:beep_factor=4:duration=5
4292 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4295 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4299 @c man end AUDIO SOURCES
4301 @chapter Audio Sinks
4302 @c man begin AUDIO SINKS
4304 Below is a description of the currently available audio sinks.
4306 @section abuffersink
4308 Buffer audio frames, and make them available to the end of filter chain.
4310 This sink is mainly intended for programmatic use, in particular
4311 through the interface defined in @file{libavfilter/buffersink.h}
4312 or the options system.
4314 It accepts a pointer to an AVABufferSinkContext structure, which
4315 defines the incoming buffers' formats, to be passed as the opaque
4316 parameter to @code{avfilter_init_filter} for initialization.
4319 Null audio sink; do absolutely nothing with the input audio. It is
4320 mainly useful as a template and for use in analysis / debugging
4323 @c man end AUDIO SINKS
4325 @chapter Video Filters
4326 @c man begin VIDEO FILTERS
4328 When you configure your FFmpeg build, you can disable any of the
4329 existing filters using @code{--disable-filters}.
4330 The configure output will show the video filters included in your
4333 Below is a description of the currently available video filters.
4335 @section alphaextract
4337 Extract the alpha component from the input as a grayscale video. This
4338 is especially useful with the @var{alphamerge} filter.
4342 Add or replace the alpha component of the primary input with the
4343 grayscale value of a second input. This is intended for use with
4344 @var{alphaextract} to allow the transmission or storage of frame
4345 sequences that have alpha in a format that doesn't support an alpha
4348 For example, to reconstruct full frames from a normal YUV-encoded video
4349 and a separate video created with @var{alphaextract}, you might use:
4351 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4354 Since this filter is designed for reconstruction, it operates on frame
4355 sequences without considering timestamps, and terminates when either
4356 input reaches end of stream. This will cause problems if your encoding
4357 pipeline drops frames. If you're trying to apply an image as an
4358 overlay to a video stream, consider the @var{overlay} filter instead.
4362 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4363 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4364 Substation Alpha) subtitles files.
4366 This filter accepts the following option in addition to the common options from
4367 the @ref{subtitles} filter:
4371 Set the shaping engine
4373 Available values are:
4376 The default libass shaping engine, which is the best available.
4378 Fast, font-agnostic shaper that can do only substitutions
4380 Slower shaper using OpenType for substitutions and positioning
4383 The default is @code{auto}.
4387 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4389 The filter accepts the following options:
4393 Set threshold A for 1st plane. Default is 0.02.
4394 Valid range is 0 to 0.3.
4397 Set threshold B for 1st plane. Default is 0.04.
4398 Valid range is 0 to 5.
4401 Set threshold A for 2nd plane. Default is 0.02.
4402 Valid range is 0 to 0.3.
4405 Set threshold B for 2nd plane. Default is 0.04.
4406 Valid range is 0 to 5.
4409 Set threshold A for 3rd plane. Default is 0.02.
4410 Valid range is 0 to 0.3.
4413 Set threshold B for 3rd plane. Default is 0.04.
4414 Valid range is 0 to 5.
4416 Threshold A is designed to react on abrupt changes in the input signal and
4417 threshold B is designed to react on continuous changes in the input signal.
4420 Set number of frames filter will use for averaging. Default is 33. Must be odd
4421 number in range [5, 129].
4424 Set what planes of frame filter will use for averaging. Default is all.
4429 Apply average blur filter.
4431 The filter accepts the following options:
4435 Set horizontal kernel size.
4438 Set which planes to filter. By default all planes are filtered.
4441 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4442 Default is @code{0}.
4447 Compute the bounding box for the non-black pixels in the input frame
4450 This filter computes the bounding box containing all the pixels with a
4451 luminance value greater than the minimum allowed value.
4452 The parameters describing the bounding box are printed on the filter
4455 The filter accepts the following option:
4459 Set the minimal luminance value. Default is @code{16}.
4462 @section bitplanenoise
4464 Show and measure bit plane noise.
4466 The filter accepts the following options:
4470 Set which plane to analyze. Default is @code{1}.
4473 Filter out noisy pixels from @code{bitplane} set above.
4474 Default is disabled.
4477 @section blackdetect
4479 Detect video intervals that are (almost) completely black. Can be
4480 useful to detect chapter transitions, commercials, or invalid
4481 recordings. Output lines contains the time for the start, end and
4482 duration of the detected black interval expressed in seconds.
4484 In order to display the output lines, you need to set the loglevel at
4485 least to the AV_LOG_INFO value.
4487 The filter accepts the following options:
4490 @item black_min_duration, d
4491 Set the minimum detected black duration expressed in seconds. It must
4492 be a non-negative floating point number.
4494 Default value is 2.0.
4496 @item picture_black_ratio_th, pic_th
4497 Set the threshold for considering a picture "black".
4498 Express the minimum value for the ratio:
4500 @var{nb_black_pixels} / @var{nb_pixels}
4503 for which a picture is considered black.
4504 Default value is 0.98.
4506 @item pixel_black_th, pix_th
4507 Set the threshold for considering a pixel "black".
4509 The threshold expresses the maximum pixel luminance value for which a
4510 pixel is considered "black". The provided value is scaled according to
4511 the following equation:
4513 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4516 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4517 the input video format, the range is [0-255] for YUV full-range
4518 formats and [16-235] for YUV non full-range formats.
4520 Default value is 0.10.
4523 The following example sets the maximum pixel threshold to the minimum
4524 value, and detects only black intervals of 2 or more seconds:
4526 blackdetect=d=2:pix_th=0.00
4531 Detect frames that are (almost) completely black. Can be useful to
4532 detect chapter transitions or commercials. Output lines consist of
4533 the frame number of the detected frame, the percentage of blackness,
4534 the position in the file if known or -1 and the timestamp in seconds.
4536 In order to display the output lines, you need to set the loglevel at
4537 least to the AV_LOG_INFO value.
4539 This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4540 The value represents the percentage of pixels in the picture that
4541 are below the threshold value.
4543 It accepts the following parameters:
4548 The percentage of the pixels that have to be below the threshold; it defaults to
4551 @item threshold, thresh
4552 The threshold below which a pixel value is considered black; it defaults to
4557 @section blend, tblend
4559 Blend two video frames into each other.
4561 The @code{blend} filter takes two input streams and outputs one
4562 stream, the first input is the "top" layer and second input is
4563 "bottom" layer. By default, the output terminates when the longest input terminates.
4565 The @code{tblend} (time blend) filter takes two consecutive frames
4566 from one single stream, and outputs the result obtained by blending
4567 the new frame on top of the old frame.
4569 A description of the accepted options follows.
4577 Set blend mode for specific pixel component or all pixel components in case
4578 of @var{all_mode}. Default value is @code{normal}.
4580 Available values for component modes are:
4621 Set blend opacity for specific pixel component or all pixel components in case
4622 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4629 Set blend expression for specific pixel component or all pixel components in case
4630 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4632 The expressions can use the following variables:
4636 The sequential number of the filtered frame, starting from @code{0}.
4640 the coordinates of the current sample
4644 the width and height of currently filtered plane
4648 Width and height scale depending on the currently filtered plane. It is the
4649 ratio between the corresponding luma plane number of pixels and the current
4650 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4651 @code{0.5,0.5} for chroma planes.
4654 Time of the current frame, expressed in seconds.
4657 Value of pixel component at current location for first video frame (top layer).
4660 Value of pixel component at current location for second video frame (bottom layer).
4664 Force termination when the shortest input terminates. Default is
4665 @code{0}. This option is only defined for the @code{blend} filter.
4668 Continue applying the last bottom frame after the end of the stream. A value of
4669 @code{0} disable the filter after the last frame of the bottom layer is reached.
4670 Default is @code{1}. This option is only defined for the @code{blend} filter.
4673 @subsection Examples
4677 Apply transition from bottom layer to top layer in first 10 seconds:
4679 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4683 Apply 1x1 checkerboard effect:
4685 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4689 Apply uncover left effect:
4691 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4695 Apply uncover down effect:
4697 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4701 Apply uncover up-left effect:
4703 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4707 Split diagonally video and shows top and bottom layer on each side:
4709 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4713 Display differences between the current and the previous frame:
4715 tblend=all_mode=difference128
4721 Apply a boxblur algorithm to the input video.
4723 It accepts the following parameters:
4727 @item luma_radius, lr
4728 @item luma_power, lp
4729 @item chroma_radius, cr
4730 @item chroma_power, cp
4731 @item alpha_radius, ar
4732 @item alpha_power, ap
4736 A description of the accepted options follows.
4739 @item luma_radius, lr
4740 @item chroma_radius, cr
4741 @item alpha_radius, ar
4742 Set an expression for the box radius in pixels used for blurring the
4743 corresponding input plane.
4745 The radius value must be a non-negative number, and must not be
4746 greater than the value of the expression @code{min(w,h)/2} for the
4747 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4750 Default value for @option{luma_radius} is "2". If not specified,
4751 @option{chroma_radius} and @option{alpha_radius} default to the
4752 corresponding value set for @option{luma_radius}.
4754 The expressions can contain the following constants:
4758 The input width and height in pixels.
4762 The input chroma image width and height in pixels.
4766 The horizontal and vertical chroma subsample values. For example, for the
4767 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4770 @item luma_power, lp
4771 @item chroma_power, cp
4772 @item alpha_power, ap
4773 Specify how many times the boxblur filter is applied to the
4774 corresponding plane.
4776 Default value for @option{luma_power} is 2. If not specified,
4777 @option{chroma_power} and @option{alpha_power} default to the
4778 corresponding value set for @option{luma_power}.
4780 A value of 0 will disable the effect.
4783 @subsection Examples
4787 Apply a boxblur filter with the luma, chroma, and alpha radii
4790 boxblur=luma_radius=2:luma_power=1
4795 Set the luma radius to 2, and alpha and chroma radius to 0:
4797 boxblur=2:1:cr=0:ar=0
4801 Set the luma and chroma radii to a fraction of the video dimension:
4803 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4809 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4810 Deinterlacing Filter").
4812 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4813 interpolation algorithms.
4814 It accepts the following parameters:
4818 The interlacing mode to adopt. It accepts one of the following values:
4822 Output one frame for each frame.
4824 Output one frame for each field.
4827 The default value is @code{send_field}.
4830 The picture field parity assumed for the input interlaced video. It accepts one
4831 of the following values:
4835 Assume the top field is first.
4837 Assume the bottom field is first.
4839 Enable automatic detection of field parity.
4842 The default value is @code{auto}.
4843 If the interlacing is unknown or the decoder does not export this information,
4844 top field first will be assumed.
4847 Specify which frames to deinterlace. Accept one of the following
4852 Deinterlace all frames.
4854 Only deinterlace frames marked as interlaced.
4857 The default value is @code{all}.
4861 YUV colorspace color/chroma keying.
4863 The filter accepts the following options:
4867 The color which will be replaced with transparency.
4870 Similarity percentage with the key color.
4872 0.01 matches only the exact key color, while 1.0 matches everything.
4877 0.0 makes pixels either fully transparent, or not transparent at all.
4879 Higher values result in semi-transparent pixels, with a higher transparency
4880 the more similar the pixels color is to the key color.
4883 Signals that the color passed is already in YUV instead of RGB.
4885 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4886 This can be used to pass exact YUV values as hexadecimal numbers.
4889 @subsection Examples
4893 Make every green pixel in the input image transparent:
4895 ffmpeg -i input.png -vf chromakey=green out.png
4899 Overlay a greenscreen-video on top of a static black background.
4901 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
4907 Display CIE color diagram with pixels overlaid onto it.
4909 The filter accepts the following options:
4924 @item uhdtv, rec2020
4937 Set what gamuts to draw.
4939 See @code{system} option for available values.
4942 Set ciescope size, by default set to 512.
4945 Set intensity used to map input pixel values to CIE diagram.
4948 Set contrast used to draw tongue colors that are out of active color system gamut.
4951 Correct gamma displayed on scope, by default enabled.
4954 Show white point on CIE diagram, by default disabled.
4957 Set input gamma. Used only with XYZ input color space.
4962 Visualize information exported by some codecs.
4964 Some codecs can export information through frames using side-data or other
4965 means. For example, some MPEG based codecs export motion vectors through the
4966 @var{export_mvs} flag in the codec @option{flags2} option.
4968 The filter accepts the following option:
4972 Set motion vectors to visualize.
4974 Available flags for @var{mv} are:
4978 forward predicted MVs of P-frames
4980 forward predicted MVs of B-frames
4982 backward predicted MVs of B-frames
4986 Display quantization parameters using the chroma planes.
4989 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4991 Available flags for @var{mv_type} are:
4995 forward predicted MVs
4997 backward predicted MVs
5000 @item frame_type, ft
5001 Set frame type to visualize motion vectors of.
5003 Available flags for @var{frame_type} are:
5007 intra-coded frames (I-frames)
5009 predicted frames (P-frames)
5011 bi-directionally predicted frames (B-frames)
5015 @subsection Examples
5019 Visualize forward predicted MVs of all frames using @command{ffplay}:
5021 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
5025 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5027 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5031 @section colorbalance
5032 Modify intensity of primary colors (red, green and blue) of input frames.
5034 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5035 regions for the red-cyan, green-magenta or blue-yellow balance.
5037 A positive adjustment value shifts the balance towards the primary color, a negative
5038 value towards the complementary color.
5040 The filter accepts the following options:
5046 Adjust red, green and blue shadows (darkest pixels).
5051 Adjust red, green and blue midtones (medium pixels).
5056 Adjust red, green and blue highlights (brightest pixels).
5058 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5061 @subsection Examples
5065 Add red color cast to shadows:
5072 RGB colorspace color keying.
5074 The filter accepts the following options:
5078 The color which will be replaced with transparency.
5081 Similarity percentage with the key color.
5083 0.01 matches only the exact key color, while 1.0 matches everything.
5088 0.0 makes pixels either fully transparent, or not transparent at all.
5090 Higher values result in semi-transparent pixels, with a higher transparency
5091 the more similar the pixels color is to the key color.
5094 @subsection Examples
5098 Make every green pixel in the input image transparent:
5100 ffmpeg -i input.png -vf colorkey=green out.png
5104 Overlay a greenscreen-video on top of a static background image.
5106 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
5110 @section colorlevels
5112 Adjust video input frames using levels.
5114 The filter accepts the following options:
5121 Adjust red, green, blue and alpha input black point.
5122 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5128 Adjust red, green, blue and alpha input white point.
5129 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
5131 Input levels are used to lighten highlights (bright tones), darken shadows
5132 (dark tones), change the balance of bright and dark tones.
5138 Adjust red, green, blue and alpha output black point.
5139 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5145 Adjust red, green, blue and alpha output white point.
5146 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5148 Output levels allows manual selection of a constrained output level range.
5151 @subsection Examples
5155 Make video output darker:
5157 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5163 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5167 Make video output lighter:
5169 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5173 Increase brightness:
5175 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5179 @section colorchannelmixer
5181 Adjust video input frames by re-mixing color channels.
5183 This filter modifies a color channel by adding the values associated to
5184 the other channels of the same pixels. For example if the value to
5185 modify is red, the output value will be:
5187 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5190 The filter accepts the following options:
5197 Adjust contribution of input red, green, blue and alpha channels for output red channel.
5198 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5204 Adjust contribution of input red, green, blue and alpha channels for output green channel.
5205 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5211 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5212 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5218 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5219 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5221 Allowed ranges for options are @code{[-2.0, 2.0]}.
5224 @subsection Examples
5228 Convert source to grayscale:
5230 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5233 Simulate sepia tones:
5235 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5239 @section colormatrix
5241 Convert color matrix.
5243 The filter accepts the following options:
5248 Specify the source and destination color matrix. Both values must be
5251 The accepted values are:
5279 For example to convert from BT.601 to SMPTE-240M, use the command:
5281 colormatrix=bt601:smpte240m
5286 Convert colorspace, transfer characteristics or color primaries.
5287 Input video needs to have an even size.
5289 The filter accepts the following options:
5294 Specify all color properties at once.
5296 The accepted values are:
5326 Specify output colorspace.
5328 The accepted values are:
5337 BT.470BG or BT.601-6 625
5340 SMPTE-170M or BT.601-6 525
5349 BT.2020 with non-constant luminance
5355 Specify output transfer characteristics.
5357 The accepted values are:
5369 Constant gamma of 2.2
5372 Constant gamma of 2.8
5375 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5393 BT.2020 for 10-bits content
5396 BT.2020 for 12-bits content
5402 Specify output color primaries.
5404 The accepted values are:
5413 BT.470BG or BT.601-6 625
5416 SMPTE-170M or BT.601-6 525
5437 Specify output color range.
5439 The accepted values are:
5442 TV (restricted) range
5445 MPEG (restricted) range
5456 Specify output color format.
5458 The accepted values are:
5461 YUV 4:2:0 planar 8-bits
5464 YUV 4:2:0 planar 10-bits
5467 YUV 4:2:0 planar 12-bits
5470 YUV 4:2:2 planar 8-bits
5473 YUV 4:2:2 planar 10-bits
5476 YUV 4:2:2 planar 12-bits
5479 YUV 4:4:4 planar 8-bits
5482 YUV 4:4:4 planar 10-bits
5485 YUV 4:4:4 planar 12-bits
5490 Do a fast conversion, which skips gamma/primary correction. This will take
5491 significantly less CPU, but will be mathematically incorrect. To get output
5492 compatible with that produced by the colormatrix filter, use fast=1.
5495 Specify dithering mode.
5497 The accepted values are:
5503 Floyd-Steinberg dithering
5507 Whitepoint adaptation mode.
5509 The accepted values are:
5512 Bradford whitepoint adaptation
5515 von Kries whitepoint adaptation
5518 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5522 Override all input properties at once. Same accepted values as @ref{all}.
5525 Override input colorspace. Same accepted values as @ref{space}.
5528 Override input color primaries. Same accepted values as @ref{primaries}.
5531 Override input transfer characteristics. Same accepted values as @ref{trc}.
5534 Override input color range. Same accepted values as @ref{range}.
5538 The filter converts the transfer characteristics, color space and color
5539 primaries to the specified user values. The output value, if not specified,
5540 is set to a default value based on the "all" property. If that property is
5541 also not specified, the filter will log an error. The output color range and
5542 format default to the same value as the input color range and format. The
5543 input transfer characteristics, color space, color primaries and color range
5544 should be set on the input data. If any of these are missing, the filter will
5545 log an error and no conversion will take place.
5547 For example to convert the input to SMPTE-240M, use the command:
5549 colorspace=smpte240m
5552 @section convolution
5554 Apply convolution 3x3 or 5x5 filter.
5556 The filter accepts the following options:
5563 Set matrix for each plane.
5564 Matrix is sequence of 9 or 25 signed integers.
5570 Set multiplier for calculated value for each plane.
5576 Set bias for each plane. This value is added to the result of the multiplication.
5577 Useful for making the overall image brighter or darker. Default is 0.0.
5580 @subsection Examples
5586 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"
5592 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"
5598 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"
5604 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"
5610 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"
5616 Copy the input source unchanged to the output. This is mainly useful for
5621 Video filtering on GPU using Apple's CoreImage API on OSX.
5623 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5624 processed by video hardware. However, software-based OpenGL implementations
5625 exist which means there is no guarantee for hardware processing. It depends on
5628 There are many filters and image generators provided by Apple that come with a
5629 large variety of options. The filter has to be referenced by its name along
5632 The coreimage filter accepts the following options:
5635 List all available filters and generators along with all their respective
5636 options as well as possible minimum and maximum values along with the default
5643 Specify all filters by their respective name and options.
5644 Use @var{list_filters} to determine all valid filter names and options.
5645 Numerical options are specified by a float value and are automatically clamped
5646 to their respective value range. Vector and color options have to be specified
5647 by a list of space separated float values. Character escaping has to be done.
5648 A special option name @code{default} is available to use default options for a
5651 It is required to specify either @code{default} or at least one of the filter options.
5652 All omitted options are used with their default values.
5653 The syntax of the filter string is as follows:
5655 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5659 Specify a rectangle where the output of the filter chain is copied into the
5660 input image. It is given by a list of space separated float values:
5662 output_rect=x\ y\ width\ height
5664 If not given, the output rectangle equals the dimensions of the input image.
5665 The output rectangle is automatically cropped at the borders of the input
5666 image. Negative values are valid for each component.
5668 output_rect=25\ 25\ 100\ 100
5672 Several filters can be chained for successive processing without GPU-HOST
5673 transfers allowing for fast processing of complex filter chains.
5674 Currently, only filters with zero (generators) or exactly one (filters) input
5675 image and one output image are supported. Also, transition filters are not yet
5678 Some filters generate output images with additional padding depending on the
5679 respective filter kernel. The padding is automatically removed to ensure the
5680 filter output has the same size as the input image.
5682 For image generators, the size of the output image is determined by the
5683 previous output image of the filter chain or the input image of the whole
5684 filterchain, respectively. The generators do not use the pixel information of
5685 this image to generate their output. However, the generated output is
5686 blended onto this image, resulting in partial or complete coverage of the
5689 The @ref{coreimagesrc} video source can be used for generating input images
5690 which are directly fed into the filter chain. By using it, providing input
5691 images by another video source or an input video is not required.
5693 @subsection Examples
5698 List all filters available:
5700 coreimage=list_filters=true
5704 Use the CIBoxBlur filter with default options to blur an image:
5706 coreimage=filter=CIBoxBlur@@default
5710 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5711 its center at 100x100 and a radius of 50 pixels:
5713 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5717 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5718 given as complete and escaped command-line for Apple's standard bash shell:
5720 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5726 Crop the input video to given dimensions.
5728 It accepts the following parameters:
5732 The width of the output video. It defaults to @code{iw}.
5733 This expression is evaluated only once during the filter
5734 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5737 The height of the output video. It defaults to @code{ih}.
5738 This expression is evaluated only once during the filter
5739 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5742 The horizontal position, in the input video, of the left edge of the output
5743 video. It defaults to @code{(in_w-out_w)/2}.
5744 This expression is evaluated per-frame.
5747 The vertical position, in the input video, of the top edge of the output video.
5748 It defaults to @code{(in_h-out_h)/2}.
5749 This expression is evaluated per-frame.
5752 If set to 1 will force the output display aspect ratio
5753 to be the same of the input, by changing the output sample aspect
5754 ratio. It defaults to 0.
5757 Enable exact cropping. If enabled, subsampled videos will be cropped at exact
5758 width/height/x/y as specified and will not be rounded to nearest smaller value.
5762 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5763 expressions containing the following constants:
5768 The computed values for @var{x} and @var{y}. They are evaluated for
5773 The input width and height.
5777 These are the same as @var{in_w} and @var{in_h}.
5781 The output (cropped) width and height.
5785 These are the same as @var{out_w} and @var{out_h}.
5788 same as @var{iw} / @var{ih}
5791 input sample aspect ratio
5794 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5798 horizontal and vertical chroma subsample values. For example for the
5799 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5802 The number of the input frame, starting from 0.
5805 the position in the file of the input frame, NAN if unknown
5808 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5812 The expression for @var{out_w} may depend on the value of @var{out_h},
5813 and the expression for @var{out_h} may depend on @var{out_w}, but they
5814 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5815 evaluated after @var{out_w} and @var{out_h}.
5817 The @var{x} and @var{y} parameters specify the expressions for the
5818 position of the top-left corner of the output (non-cropped) area. They
5819 are evaluated for each frame. If the evaluated value is not valid, it
5820 is approximated to the nearest valid value.
5822 The expression for @var{x} may depend on @var{y}, and the expression
5823 for @var{y} may depend on @var{x}.
5825 @subsection Examples
5829 Crop area with size 100x100 at position (12,34).
5834 Using named options, the example above becomes:
5836 crop=w=100:h=100:x=12:y=34
5840 Crop the central input area with size 100x100:
5846 Crop the central input area with size 2/3 of the input video:
5848 crop=2/3*in_w:2/3*in_h
5852 Crop the input video central square:
5859 Delimit the rectangle with the top-left corner placed at position
5860 100:100 and the right-bottom corner corresponding to the right-bottom
5861 corner of the input image.
5863 crop=in_w-100:in_h-100:100:100
5867 Crop 10 pixels from the left and right borders, and 20 pixels from
5868 the top and bottom borders
5870 crop=in_w-2*10:in_h-2*20
5874 Keep only the bottom right quarter of the input image:
5876 crop=in_w/2:in_h/2:in_w/2:in_h/2
5880 Crop height for getting Greek harmony:
5882 crop=in_w:1/PHI*in_w
5886 Apply trembling effect:
5888 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)
5892 Apply erratic camera effect depending on timestamp:
5894 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)"
5898 Set x depending on the value of y:
5900 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5904 @subsection Commands
5906 This filter supports the following commands:
5912 Set width/height of the output video and the horizontal/vertical position
5914 The command accepts the same syntax of the corresponding option.
5916 If the specified expression is not valid, it is kept at its current
5922 Auto-detect the crop size.
5924 It calculates the necessary cropping parameters and prints the
5925 recommended parameters via the logging system. The detected dimensions
5926 correspond to the non-black area of the input video.
5928 It accepts the following parameters:
5933 Set higher black value threshold, which can be optionally specified
5934 from nothing (0) to everything (255 for 8-bit based formats). An intensity
5935 value greater to the set value is considered non-black. It defaults to 24.
5936 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5937 on the bitdepth of the pixel format.
5940 The value which the width/height should be divisible by. It defaults to
5941 16. The offset is automatically adjusted to center the video. Use 2 to
5942 get only even dimensions (needed for 4:2:2 video). 16 is best when
5943 encoding to most video codecs.
5945 @item reset_count, reset
5946 Set the counter that determines after how many frames cropdetect will
5947 reset the previously detected largest video area and start over to
5948 detect the current optimal crop area. Default value is 0.
5950 This can be useful when channel logos distort the video area. 0
5951 indicates 'never reset', and returns the largest area encountered during
5958 Apply color adjustments using curves.
5960 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5961 component (red, green and blue) has its values defined by @var{N} key points
5962 tied from each other using a smooth curve. The x-axis represents the pixel
5963 values from the input frame, and the y-axis the new pixel values to be set for
5966 By default, a component curve is defined by the two points @var{(0;0)} and
5967 @var{(1;1)}. This creates a straight line where each original pixel value is
5968 "adjusted" to its own value, which means no change to the image.
5970 The filter allows you to redefine these two points and add some more. A new
5971 curve (using a natural cubic spline interpolation) will be define to pass
5972 smoothly through all these new coordinates. The new defined points needs to be
5973 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5974 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5975 the vector spaces, the values will be clipped accordingly.
5977 The filter accepts the following options:
5981 Select one of the available color presets. This option can be used in addition
5982 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5983 options takes priority on the preset values.
5984 Available presets are:
5987 @item color_negative
5990 @item increase_contrast
5992 @item linear_contrast
5993 @item medium_contrast
5995 @item strong_contrast
5998 Default is @code{none}.
6000 Set the master key points. These points will define a second pass mapping. It
6001 is sometimes called a "luminance" or "value" mapping. It can be used with
6002 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
6003 post-processing LUT.
6005 Set the key points for the red component.
6007 Set the key points for the green component.
6009 Set the key points for the blue component.
6011 Set the key points for all components (not including master).
6012 Can be used in addition to the other key points component
6013 options. In this case, the unset component(s) will fallback on this
6014 @option{all} setting.
6016 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
6018 Save Gnuplot script of the curves in specified file.
6021 To avoid some filtergraph syntax conflicts, each key points list need to be
6022 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
6024 @subsection Examples
6028 Increase slightly the middle level of blue:
6030 curves=blue='0/0 0.5/0.58 1/1'
6036 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'
6038 Here we obtain the following coordinates for each components:
6041 @code{(0;0.11) (0.42;0.51) (1;0.95)}
6043 @code{(0;0) (0.50;0.48) (1;1)}
6045 @code{(0;0.22) (0.49;0.44) (1;0.80)}
6049 The previous example can also be achieved with the associated built-in preset:
6051 curves=preset=vintage
6061 Use a Photoshop preset and redefine the points of the green component:
6063 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6067 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6068 and @command{gnuplot}:
6070 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
6071 gnuplot -p /tmp/curves.plt
6077 Video data analysis filter.
6079 This filter shows hexadecimal pixel values of part of video.
6081 The filter accepts the following options:
6085 Set output video size.
6088 Set x offset from where to pick pixels.
6091 Set y offset from where to pick pixels.
6094 Set scope mode, can be one of the following:
6097 Draw hexadecimal pixel values with white color on black background.
6100 Draw hexadecimal pixel values with input video pixel color on black
6104 Draw hexadecimal pixel values on color background picked from input video,
6105 the text color is picked in such way so its always visible.
6109 Draw rows and columns numbers on left and top of video.
6112 Set background opacity.
6117 Denoise frames using 2D DCT (frequency domain filtering).
6119 This filter is not designed for real time.
6121 The filter accepts the following options:
6125 Set the noise sigma constant.
6127 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6128 coefficient (absolute value) below this threshold with be dropped.
6130 If you need a more advanced filtering, see @option{expr}.
6132 Default is @code{0}.
6135 Set number overlapping pixels for each block. Since the filter can be slow, you
6136 may want to reduce this value, at the cost of a less effective filter and the
6137 risk of various artefacts.
6139 If the overlapping value doesn't permit processing the whole input width or
6140 height, a warning will be displayed and according borders won't be denoised.
6142 Default value is @var{blocksize}-1, which is the best possible setting.
6145 Set the coefficient factor expression.
6147 For each coefficient of a DCT block, this expression will be evaluated as a
6148 multiplier value for the coefficient.
6150 If this is option is set, the @option{sigma} option will be ignored.
6152 The absolute value of the coefficient can be accessed through the @var{c}
6156 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
6157 @var{blocksize}, which is the width and height of the processed blocks.
6159 The default value is @var{3} (8x8) and can be raised to @var{4} for a
6160 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
6161 on the speed processing. Also, a larger block size does not necessarily means a
6165 @subsection Examples
6167 Apply a denoise with a @option{sigma} of @code{4.5}:
6172 The same operation can be achieved using the expression system:
6174 dctdnoiz=e='gte(c, 4.5*3)'
6177 Violent denoise using a block size of @code{16x16}:
6184 Remove banding artifacts from input video.
6185 It works by replacing banded pixels with average value of referenced pixels.
6187 The filter accepts the following options:
6194 Set banding detection threshold for each plane. Default is 0.02.
6195 Valid range is 0.00003 to 0.5.
6196 If difference between current pixel and reference pixel is less than threshold,
6197 it will be considered as banded.
6200 Banding detection range in pixels. Default is 16. If positive, random number
6201 in range 0 to set value will be used. If negative, exact absolute value
6203 The range defines square of four pixels around current pixel.
6206 Set direction in radians from which four pixel will be compared. If positive,
6207 random direction from 0 to set direction will be picked. If negative, exact of
6208 absolute value will be picked. For example direction 0, -PI or -2*PI radians
6209 will pick only pixels on same row and -PI/2 will pick only pixels on same
6213 If enabled, current pixel is compared with average value of all four
6214 surrounding pixels. The default is enabled. If disabled current pixel is
6215 compared with all four surrounding pixels. The pixel is considered banded
6216 if only all four differences with surrounding pixels are less than threshold.
6219 If enabled, current pixel is changed if and only if all pixel components are banded,
6220 e.g. banding detection threshold is triggered for all color components.
6221 The default is disabled.
6227 Drop duplicated frames at regular intervals.
6229 The filter accepts the following options:
6233 Set the number of frames from which one will be dropped. Setting this to
6234 @var{N} means one frame in every batch of @var{N} frames will be dropped.
6235 Default is @code{5}.
6238 Set the threshold for duplicate detection. If the difference metric for a frame
6239 is less than or equal to this value, then it is declared as duplicate. Default
6243 Set scene change threshold. Default is @code{15}.
6247 Set the size of the x and y-axis blocks used during metric calculations.
6248 Larger blocks give better noise suppression, but also give worse detection of
6249 small movements. Must be a power of two. Default is @code{32}.
6252 Mark main input as a pre-processed input and activate clean source input
6253 stream. This allows the input to be pre-processed with various filters to help
6254 the metrics calculation while keeping the frame selection lossless. When set to
6255 @code{1}, the first stream is for the pre-processed input, and the second
6256 stream is the clean source from where the kept frames are chosen. Default is
6260 Set whether or not chroma is considered in the metric calculations. Default is
6266 Apply deflate effect to the video.
6268 This filter replaces the pixel by the local(3x3) average by taking into account
6269 only values lower than the pixel.
6271 It accepts the following options:
6278 Limit the maximum change for each plane, default is 65535.
6279 If 0, plane will remain unchanged.
6284 Remove temporal frame luminance variations.
6286 It accepts the following options:
6290 Set moving-average filter size in frames. Default is 5. Allowed range is 2 - 129.
6293 Set averaging mode to smooth temporal luminance variations.
6295 Available values are:
6322 Remove judder produced by partially interlaced telecined content.
6324 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6325 source was partially telecined content then the output of @code{pullup,dejudder}
6326 will have a variable frame rate. May change the recorded frame rate of the
6327 container. Aside from that change, this filter will not affect constant frame
6330 The option available in this filter is:
6334 Specify the length of the window over which the judder repeats.
6336 Accepts any integer greater than 1. Useful values are:
6340 If the original was telecined from 24 to 30 fps (Film to NTSC).
6343 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6346 If a mixture of the two.
6349 The default is @samp{4}.
6354 Suppress a TV station logo by a simple interpolation of the surrounding
6355 pixels. Just set a rectangle covering the logo and watch it disappear
6356 (and sometimes something even uglier appear - your mileage may vary).
6358 It accepts the following parameters:
6363 Specify the top left corner coordinates of the logo. They must be
6368 Specify the width and height of the logo to clear. They must be
6372 Specify the thickness of the fuzzy edge of the rectangle (added to
6373 @var{w} and @var{h}). The default value is 1. This option is
6374 deprecated, setting higher values should no longer be necessary and
6378 When set to 1, a green rectangle is drawn on the screen to simplify
6379 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6380 The default value is 0.
6382 The rectangle is drawn on the outermost pixels which will be (partly)
6383 replaced with interpolated values. The values of the next pixels
6384 immediately outside this rectangle in each direction will be used to
6385 compute the interpolated pixel values inside the rectangle.
6389 @subsection Examples
6393 Set a rectangle covering the area with top left corner coordinates 0,0
6394 and size 100x77, and a band of size 10:
6396 delogo=x=0:y=0:w=100:h=77:band=10
6403 Attempt to fix small changes in horizontal and/or vertical shift. This
6404 filter helps remove camera shake from hand-holding a camera, bumping a
6405 tripod, moving on a vehicle, etc.
6407 The filter accepts the following options:
6415 Specify a rectangular area where to limit the search for motion
6417 If desired the search for motion vectors can be limited to a
6418 rectangular area of the frame defined by its top left corner, width
6419 and height. These parameters have the same meaning as the drawbox
6420 filter which can be used to visualise the position of the bounding
6423 This is useful when simultaneous movement of subjects within the frame
6424 might be confused for camera motion by the motion vector search.
6426 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6427 then the full frame is used. This allows later options to be set
6428 without specifying the bounding box for the motion vector search.
6430 Default - search the whole frame.
6434 Specify the maximum extent of movement in x and y directions in the
6435 range 0-64 pixels. Default 16.
6438 Specify how to generate pixels to fill blanks at the edge of the
6439 frame. Available values are:
6442 Fill zeroes at blank locations
6444 Original image at blank locations
6446 Extruded edge value at blank locations
6448 Mirrored edge at blank locations
6450 Default value is @samp{mirror}.
6453 Specify the blocksize to use for motion search. Range 4-128 pixels,
6457 Specify the contrast threshold for blocks. Only blocks with more than
6458 the specified contrast (difference between darkest and lightest
6459 pixels) will be considered. Range 1-255, default 125.
6462 Specify the search strategy. Available values are:
6465 Set exhaustive search
6467 Set less exhaustive search.
6469 Default value is @samp{exhaustive}.
6472 If set then a detailed log of the motion search is written to the
6476 If set to 1, specify using OpenCL capabilities, only available if
6477 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6483 Apply an exact inverse of the telecine operation. It requires a predefined
6484 pattern specified using the pattern option which must be the same as that passed
6485 to the telecine filter.
6487 This filter accepts the following options:
6496 The default value is @code{top}.
6500 A string of numbers representing the pulldown pattern you wish to apply.
6501 The default value is @code{23}.
6504 A number representing position of the first frame with respect to the telecine
6505 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6510 Apply dilation effect to the video.
6512 This filter replaces the pixel by the local(3x3) maximum.
6514 It accepts the following options:
6521 Limit the maximum change for each plane, default is 65535.
6522 If 0, plane will remain unchanged.
6525 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6528 Flags to local 3x3 coordinates maps like this:
6537 Displace pixels as indicated by second and third input stream.
6539 It takes three input streams and outputs one stream, the first input is the
6540 source, and second and third input are displacement maps.
6542 The second input specifies how much to displace pixels along the
6543 x-axis, while the third input specifies how much to displace pixels
6545 If one of displacement map streams terminates, last frame from that
6546 displacement map will be used.
6548 Note that once generated, displacements maps can be reused over and over again.
6550 A description of the accepted options follows.
6554 Set displace behavior for pixels that are out of range.
6556 Available values are:
6559 Missing pixels are replaced by black pixels.
6562 Adjacent pixels will spread out to replace missing pixels.
6565 Out of range pixels are wrapped so they point to pixels of other side.
6567 Default is @samp{smear}.
6571 @subsection Examples
6575 Add ripple effect to rgb input of video size hd720:
6577 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
6581 Add wave effect to rgb input of video size hd720:
6583 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
6589 Draw a colored box on the input image.
6591 It accepts the following parameters:
6596 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6600 The expressions which specify the width and height of the box; if 0 they are interpreted as
6601 the input width and height. It defaults to 0.
6604 Specify the color of the box to write. For the general syntax of this option,
6605 check the "Color" section in the ffmpeg-utils manual. If the special
6606 value @code{invert} is used, the box edge color is the same as the
6607 video with inverted luma.
6610 The expression which sets the thickness of the box edge. Default value is @code{3}.
6612 See below for the list of accepted constants.
6615 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6616 following constants:
6620 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6624 horizontal and vertical chroma subsample values. For example for the
6625 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6629 The input width and height.
6632 The input sample aspect ratio.
6636 The x and y offset coordinates where the box is drawn.
6640 The width and height of the drawn box.
6643 The thickness of the drawn box.
6645 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6646 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6650 @subsection Examples
6654 Draw a black box around the edge of the input image:
6660 Draw a box with color red and an opacity of 50%:
6662 drawbox=10:20:200:60:red@@0.5
6665 The previous example can be specified as:
6667 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6671 Fill the box with pink color:
6673 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6677 Draw a 2-pixel red 2.40:1 mask:
6679 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
6685 Draw a grid on the input image.
6687 It accepts the following parameters:
6692 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6696 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6697 input width and height, respectively, minus @code{thickness}, so image gets
6698 framed. Default to 0.
6701 Specify the color of the grid. For the general syntax of this option,
6702 check the "Color" section in the ffmpeg-utils manual. If the special
6703 value @code{invert} is used, the grid color is the same as the
6704 video with inverted luma.
6707 The expression which sets the thickness of the grid line. Default value is @code{1}.
6709 See below for the list of accepted constants.
6712 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6713 following constants:
6717 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6721 horizontal and vertical chroma subsample values. For example for the
6722 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6726 The input grid cell width and height.
6729 The input sample aspect ratio.
6733 The x and y coordinates of some point of grid intersection (meant to configure offset).
6737 The width and height of the drawn cell.
6740 The thickness of the drawn cell.
6742 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6743 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6747 @subsection Examples
6751 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6753 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6757 Draw a white 3x3 grid with an opacity of 50%:
6759 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6766 Draw a text string or text from a specified file on top of a video, using the
6767 libfreetype library.
6769 To enable compilation of this filter, you need to configure FFmpeg with
6770 @code{--enable-libfreetype}.
6771 To enable default font fallback and the @var{font} option you need to
6772 configure FFmpeg with @code{--enable-libfontconfig}.
6773 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6774 @code{--enable-libfribidi}.
6778 It accepts the following parameters:
6783 Used to draw a box around text using the background color.
6784 The value must be either 1 (enable) or 0 (disable).
6785 The default value of @var{box} is 0.
6788 Set the width of the border to be drawn around the box using @var{boxcolor}.
6789 The default value of @var{boxborderw} is 0.
6792 The color to be used for drawing box around text. For the syntax of this
6793 option, check the "Color" section in the ffmpeg-utils manual.
6795 The default value of @var{boxcolor} is "white".
6798 Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6799 The default value of @var{line_spacing} is 0.
6802 Set the width of the border to be drawn around the text using @var{bordercolor}.
6803 The default value of @var{borderw} is 0.
6806 Set the color to be used for drawing border around text. For the syntax of this
6807 option, check the "Color" section in the ffmpeg-utils manual.
6809 The default value of @var{bordercolor} is "black".
6812 Select how the @var{text} is expanded. Can be either @code{none},
6813 @code{strftime} (deprecated) or
6814 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6818 Set a start time for the count. Value is in microseconds. Only applied
6819 in the deprecated strftime expansion mode. To emulate in normal expansion
6820 mode use the @code{pts} function, supplying the start time (in seconds)
6821 as the second argument.
6824 If true, check and fix text coords to avoid clipping.
6827 The color to be used for drawing fonts. For the syntax of this option, check
6828 the "Color" section in the ffmpeg-utils manual.
6830 The default value of @var{fontcolor} is "black".
6832 @item fontcolor_expr
6833 String which is expanded the same way as @var{text} to obtain dynamic
6834 @var{fontcolor} value. By default this option has empty value and is not
6835 processed. When this option is set, it overrides @var{fontcolor} option.
6838 The font family to be used for drawing text. By default Sans.
6841 The font file to be used for drawing text. The path must be included.
6842 This parameter is mandatory if the fontconfig support is disabled.
6845 Draw the text applying alpha blending. The value can
6846 be a number between 0.0 and 1.0.
6847 The expression accepts the same variables @var{x, y} as well.
6848 The default value is 1.
6849 Please see @var{fontcolor_expr}.
6852 The font size to be used for drawing text.
6853 The default value of @var{fontsize} is 16.
6856 If set to 1, attempt to shape the text (for example, reverse the order of
6857 right-to-left text and join Arabic characters) before drawing it.
6858 Otherwise, just draw the text exactly as given.
6859 By default 1 (if supported).
6862 The flags to be used for loading the fonts.
6864 The flags map the corresponding flags supported by libfreetype, and are
6865 a combination of the following values:
6872 @item vertical_layout
6873 @item force_autohint
6876 @item ignore_global_advance_width
6878 @item ignore_transform
6884 Default value is "default".
6886 For more information consult the documentation for the FT_LOAD_*
6890 The color to be used for drawing a shadow behind the drawn text. For the
6891 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6893 The default value of @var{shadowcolor} is "black".
6897 The x and y offsets for the text shadow position with respect to the
6898 position of the text. They can be either positive or negative
6899 values. The default value for both is "0".
6902 The starting frame number for the n/frame_num variable. The default value
6906 The size in number of spaces to use for rendering the tab.
6910 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6911 format. It can be used with or without text parameter. @var{timecode_rate}
6912 option must be specified.
6914 @item timecode_rate, rate, r
6915 Set the timecode frame rate (timecode only).
6918 If set to 1, the output of the timecode option will wrap around at 24 hours.
6919 Default is 0 (disabled).
6922 The text string to be drawn. The text must be a sequence of UTF-8
6924 This parameter is mandatory if no file is specified with the parameter
6928 A text file containing text to be drawn. The text must be a sequence
6929 of UTF-8 encoded characters.
6931 This parameter is mandatory if no text string is specified with the
6932 parameter @var{text}.
6934 If both @var{text} and @var{textfile} are specified, an error is thrown.
6937 If set to 1, the @var{textfile} will be reloaded before each frame.
6938 Be sure to update it atomically, or it may be read partially, or even fail.
6942 The expressions which specify the offsets where text will be drawn
6943 within the video frame. They are relative to the top/left border of the
6946 The default value of @var{x} and @var{y} is "0".
6948 See below for the list of accepted constants and functions.
6951 The parameters for @var{x} and @var{y} are expressions containing the
6952 following constants and functions:
6956 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6960 horizontal and vertical chroma subsample values. For example for the
6961 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6964 the height of each text line
6972 @item max_glyph_a, ascent
6973 the maximum distance from the baseline to the highest/upper grid
6974 coordinate used to place a glyph outline point, for all the rendered
6976 It is a positive value, due to the grid's orientation with the Y axis
6979 @item max_glyph_d, descent
6980 the maximum distance from the baseline to the lowest grid coordinate
6981 used to place a glyph outline point, for all the rendered glyphs.
6982 This is a negative value, due to the grid's orientation, with the Y axis
6986 maximum glyph height, that is the maximum height for all the glyphs
6987 contained in the rendered text, it is equivalent to @var{ascent} -
6991 maximum glyph width, that is the maximum width for all the glyphs
6992 contained in the rendered text
6995 the number of input frame, starting from 0
6997 @item rand(min, max)
6998 return a random number included between @var{min} and @var{max}
7001 The input sample aspect ratio.
7004 timestamp expressed in seconds, NAN if the input timestamp is unknown
7007 the height of the rendered text
7010 the width of the rendered text
7014 the x and y offset coordinates where the text is drawn.
7016 These parameters allow the @var{x} and @var{y} expressions to refer
7017 each other, so you can for example specify @code{y=x/dar}.
7020 @anchor{drawtext_expansion}
7021 @subsection Text expansion
7023 If @option{expansion} is set to @code{strftime},
7024 the filter recognizes strftime() sequences in the provided text and
7025 expands them accordingly. Check the documentation of strftime(). This
7026 feature is deprecated.
7028 If @option{expansion} is set to @code{none}, the text is printed verbatim.
7030 If @option{expansion} is set to @code{normal} (which is the default),
7031 the following expansion mechanism is used.
7033 The backslash character @samp{\}, followed by any character, always expands to
7034 the second character.
7036 Sequences of the form @code{%@{...@}} are expanded. The text between the
7037 braces is a function name, possibly followed by arguments separated by ':'.
7038 If the arguments contain special characters or delimiters (':' or '@}'),
7039 they should be escaped.
7041 Note that they probably must also be escaped as the value for the
7042 @option{text} option in the filter argument string and as the filter
7043 argument in the filtergraph description, and possibly also for the shell,
7044 that makes up to four levels of escaping; using a text file avoids these
7047 The following functions are available:
7052 The expression evaluation result.
7054 It must take one argument specifying the expression to be evaluated,
7055 which accepts the same constants and functions as the @var{x} and
7056 @var{y} values. Note that not all constants should be used, for
7057 example the text size is not known when evaluating the expression, so
7058 the constants @var{text_w} and @var{text_h} will have an undefined
7061 @item expr_int_format, eif
7062 Evaluate the expression's value and output as formatted integer.
7064 The first argument is the expression to be evaluated, just as for the @var{expr} function.
7065 The second argument specifies the output format. Allowed values are @samp{x},
7066 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7067 @code{printf} function.
7068 The third parameter is optional and sets the number of positions taken by the output.
7069 It can be used to add padding with zeros from the left.
7072 The time at which the filter is running, expressed in UTC.
7073 It can accept an argument: a strftime() format string.
7076 The time at which the filter is running, expressed in the local time zone.
7077 It can accept an argument: a strftime() format string.
7080 Frame metadata. Takes one or two arguments.
7082 The first argument is mandatory and specifies the metadata key.
7084 The second argument is optional and specifies a default value, used when the
7085 metadata key is not found or empty.
7088 The frame number, starting from 0.
7091 A 1 character description of the current picture type.
7094 The timestamp of the current frame.
7095 It can take up to three arguments.
7097 The first argument is the format of the timestamp; it defaults to @code{flt}
7098 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7099 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7100 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7101 @code{localtime} stands for the timestamp of the frame formatted as
7102 local time zone time.
7104 The second argument is an offset added to the timestamp.
7106 If the format is set to @code{localtime} or @code{gmtime},
7107 a third argument may be supplied: a strftime() format string.
7108 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7111 @subsection Examples
7115 Draw "Test Text" with font FreeSerif, using the default values for the
7116 optional parameters.
7119 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7123 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7124 and y=50 (counting from the top-left corner of the screen), text is
7125 yellow with a red box around it. Both the text and the box have an
7129 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7130 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7133 Note that the double quotes are not necessary if spaces are not used
7134 within the parameter list.
7137 Show the text at the center of the video frame:
7139 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7143 Show the text at a random position, switching to a new position every 30 seconds:
7145 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)"
7149 Show a text line sliding from right to left in the last row of the video
7150 frame. The file @file{LONG_LINE} is assumed to contain a single line
7153 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7157 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7159 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7163 Draw a single green letter "g", at the center of the input video.
7164 The glyph baseline is placed at half screen height.
7166 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7170 Show text for 1 second every 3 seconds:
7172 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7176 Use fontconfig to set the font. Note that the colons need to be escaped.
7178 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7182 Print the date of a real-time encoding (see strftime(3)):
7184 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7188 Show text fading in and out (appearing/disappearing):
7191 DS=1.0 # display start
7192 DE=10.0 # display end
7193 FID=1.5 # fade in duration
7194 FOD=5 # fade out duration
7195 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 @}"
7199 Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7200 and the @option{fontsize} value are included in the @option{y} offset.
7202 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7203 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7208 For more information about libfreetype, check:
7209 @url{http://www.freetype.org/}.
7211 For more information about fontconfig, check:
7212 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7214 For more information about libfribidi, check:
7215 @url{http://fribidi.org/}.
7219 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7221 The filter accepts the following options:
7226 Set low and high threshold values used by the Canny thresholding
7229 The high threshold selects the "strong" edge pixels, which are then
7230 connected through 8-connectivity with the "weak" edge pixels selected
7231 by the low threshold.
7233 @var{low} and @var{high} threshold values must be chosen in the range
7234 [0,1], and @var{low} should be lesser or equal to @var{high}.
7236 Default value for @var{low} is @code{20/255}, and default value for @var{high}
7240 Define the drawing mode.
7244 Draw white/gray wires on black background.
7247 Mix the colors to create a paint/cartoon effect.
7250 Default value is @var{wires}.
7253 @subsection Examples
7257 Standard edge detection with custom values for the hysteresis thresholding:
7259 edgedetect=low=0.1:high=0.4
7263 Painting effect without thresholding:
7265 edgedetect=mode=colormix:high=0
7270 Set brightness, contrast, saturation and approximate gamma adjustment.
7272 The filter accepts the following options:
7276 Set the contrast expression. The value must be a float value in range
7277 @code{-2.0} to @code{2.0}. The default value is "1".
7280 Set the brightness expression. The value must be a float value in
7281 range @code{-1.0} to @code{1.0}. The default value is "0".
7284 Set the saturation expression. The value must be a float in
7285 range @code{0.0} to @code{3.0}. The default value is "1".
7288 Set the gamma expression. The value must be a float in range
7289 @code{0.1} to @code{10.0}. The default value is "1".
7292 Set the gamma expression for red. The value must be a float in
7293 range @code{0.1} to @code{10.0}. The default value is "1".
7296 Set the gamma expression for green. The value must be a float in range
7297 @code{0.1} to @code{10.0}. The default value is "1".
7300 Set the gamma expression for blue. The value must be a float in range
7301 @code{0.1} to @code{10.0}. The default value is "1".
7304 Set the gamma weight expression. It can be used to reduce the effect
7305 of a high gamma value on bright image areas, e.g. keep them from
7306 getting overamplified and just plain white. The value must be a float
7307 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7308 gamma correction all the way down while @code{1.0} leaves it at its
7309 full strength. Default is "1".
7312 Set when the expressions for brightness, contrast, saturation and
7313 gamma expressions are evaluated.
7315 It accepts the following values:
7318 only evaluate expressions once during the filter initialization or
7319 when a command is processed
7322 evaluate expressions for each incoming frame
7325 Default value is @samp{init}.
7328 The expressions accept the following parameters:
7331 frame count of the input frame starting from 0
7334 byte position of the corresponding packet in the input file, NAN if
7338 frame rate of the input video, NAN if the input frame rate is unknown
7341 timestamp expressed in seconds, NAN if the input timestamp is unknown
7344 @subsection Commands
7345 The filter supports the following commands:
7349 Set the contrast expression.
7352 Set the brightness expression.
7355 Set the saturation expression.
7358 Set the gamma expression.
7361 Set the gamma_r expression.
7364 Set gamma_g expression.
7367 Set gamma_b expression.
7370 Set gamma_weight expression.
7372 The command accepts the same syntax of the corresponding option.
7374 If the specified expression is not valid, it is kept at its current
7381 Apply erosion effect to the video.
7383 This filter replaces the pixel by the local(3x3) minimum.
7385 It accepts the following options:
7392 Limit the maximum change for each plane, default is 65535.
7393 If 0, plane will remain unchanged.
7396 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7399 Flags to local 3x3 coordinates maps like this:
7406 @section extractplanes
7408 Extract color channel components from input video stream into
7409 separate grayscale video streams.
7411 The filter accepts the following option:
7415 Set plane(s) to extract.
7417 Available values for planes are:
7428 Choosing planes not available in the input will result in an error.
7429 That means you cannot select @code{r}, @code{g}, @code{b} planes
7430 with @code{y}, @code{u}, @code{v} planes at same time.
7433 @subsection Examples
7437 Extract luma, u and v color channel component from input video frame
7438 into 3 grayscale outputs:
7440 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
7446 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7448 For each input image, the filter will compute the optimal mapping from
7449 the input to the output given the codebook length, that is the number
7450 of distinct output colors.
7452 This filter accepts the following options.
7455 @item codebook_length, l
7456 Set codebook length. The value must be a positive integer, and
7457 represents the number of distinct output colors. Default value is 256.
7460 Set the maximum number of iterations to apply for computing the optimal
7461 mapping. The higher the value the better the result and the higher the
7462 computation time. Default value is 1.
7465 Set a random seed, must be an integer included between 0 and
7466 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7467 will try to use a good random seed on a best effort basis.
7470 Set pal8 output pixel format. This option does not work with codebook
7471 length greater than 256.
7476 Apply a fade-in/out effect to the input video.
7478 It accepts the following parameters:
7482 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7484 Default is @code{in}.
7486 @item start_frame, s
7487 Specify the number of the frame to start applying the fade
7488 effect at. Default is 0.
7491 The number of frames that the fade effect lasts. At the end of the
7492 fade-in effect, the output video will have the same intensity as the input video.
7493 At the end of the fade-out transition, the output video will be filled with the
7494 selected @option{color}.
7498 If set to 1, fade only alpha channel, if one exists on the input.
7501 @item start_time, st
7502 Specify the timestamp (in seconds) of the frame to start to apply the fade
7503 effect. If both start_frame and start_time are specified, the fade will start at
7504 whichever comes last. Default is 0.
7507 The number of seconds for which the fade effect has to last. At the end of the
7508 fade-in effect the output video will have the same intensity as the input video,
7509 at the end of the fade-out transition the output video will be filled with the
7510 selected @option{color}.
7511 If both duration and nb_frames are specified, duration is used. Default is 0
7512 (nb_frames is used by default).
7515 Specify the color of the fade. Default is "black".
7518 @subsection Examples
7522 Fade in the first 30 frames of video:
7527 The command above is equivalent to:
7533 Fade out the last 45 frames of a 200-frame video:
7536 fade=type=out:start_frame=155:nb_frames=45
7540 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7542 fade=in:0:25, fade=out:975:25
7546 Make the first 5 frames yellow, then fade in from frame 5-24:
7548 fade=in:5:20:color=yellow
7552 Fade in alpha over first 25 frames of video:
7554 fade=in:0:25:alpha=1
7558 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7560 fade=t=in:st=5.5:d=0.5
7566 Apply arbitrary expressions to samples in frequency domain
7570 Adjust the dc value (gain) of the luma plane of the image. The filter
7571 accepts an integer value in range @code{0} to @code{1000}. The default
7572 value is set to @code{0}.
7575 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7576 filter accepts an integer value in range @code{0} to @code{1000}. The
7577 default value is set to @code{0}.
7580 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7581 filter accepts an integer value in range @code{0} to @code{1000}. The
7582 default value is set to @code{0}.
7585 Set the frequency domain weight expression for the luma plane.
7588 Set the frequency domain weight expression for the 1st chroma plane.
7591 Set the frequency domain weight expression for the 2nd chroma plane.
7593 The filter accepts the following variables:
7596 The coordinates of the current sample.
7600 The width and height of the image.
7603 @subsection Examples
7609 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7615 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7621 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7627 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7634 Extract a single field from an interlaced image using stride
7635 arithmetic to avoid wasting CPU time. The output frames are marked as
7638 The filter accepts the following options:
7642 Specify whether to extract the top (if the value is @code{0} or
7643 @code{top}) or the bottom field (if the value is @code{1} or
7649 Create new frames by copying the top and bottom fields from surrounding frames
7650 supplied as numbers by the hint file.
7654 Set file containing hints: absolute/relative frame numbers.
7656 There must be one line for each frame in a clip. Each line must contain two
7657 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7658 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7659 is current frame number for @code{absolute} mode or out of [-1, 1] range
7660 for @code{relative} mode. First number tells from which frame to pick up top
7661 field and second number tells from which frame to pick up bottom field.
7663 If optionally followed by @code{+} output frame will be marked as interlaced,
7664 else if followed by @code{-} output frame will be marked as progressive, else
7665 it will be marked same as input frame.
7666 If line starts with @code{#} or @code{;} that line is skipped.
7669 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7672 Example of first several lines of @code{hint} file for @code{relative} mode:
7675 1,0 - # second frame, use third's frame top field and second's frame bottom field
7676 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7693 Field matching filter for inverse telecine. It is meant to reconstruct the
7694 progressive frames from a telecined stream. The filter does not drop duplicated
7695 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7696 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7698 The separation of the field matching and the decimation is notably motivated by
7699 the possibility of inserting a de-interlacing filter fallback between the two.
7700 If the source has mixed telecined and real interlaced content,
7701 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7702 But these remaining combed frames will be marked as interlaced, and thus can be
7703 de-interlaced by a later filter such as @ref{yadif} before decimation.
7705 In addition to the various configuration options, @code{fieldmatch} can take an
7706 optional second stream, activated through the @option{ppsrc} option. If
7707 enabled, the frames reconstruction will be based on the fields and frames from
7708 this second stream. This allows the first input to be pre-processed in order to
7709 help the various algorithms of the filter, while keeping the output lossless
7710 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7711 or brightness/contrast adjustments can help.
7713 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7714 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7715 which @code{fieldmatch} is based on. While the semantic and usage are very
7716 close, some behaviour and options names can differ.
7718 The @ref{decimate} filter currently only works for constant frame rate input.
7719 If your input has mixed telecined (30fps) and progressive content with a lower
7720 framerate like 24fps use the following filterchain to produce the necessary cfr
7721 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7723 The filter accepts the following options:
7727 Specify the assumed field order of the input stream. Available values are:
7731 Auto detect parity (use FFmpeg's internal parity value).
7733 Assume bottom field first.
7735 Assume top field first.
7738 Note that it is sometimes recommended not to trust the parity announced by the
7741 Default value is @var{auto}.
7744 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7745 sense that it won't risk creating jerkiness due to duplicate frames when
7746 possible, but if there are bad edits or blended fields it will end up
7747 outputting combed frames when a good match might actually exist. On the other
7748 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7749 but will almost always find a good frame if there is one. The other values are
7750 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7751 jerkiness and creating duplicate frames versus finding good matches in sections
7752 with bad edits, orphaned fields, blended fields, etc.
7754 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7756 Available values are:
7760 2-way matching (p/c)
7762 2-way matching, and trying 3rd match if still combed (p/c + n)
7764 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7766 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7767 still combed (p/c + n + u/b)
7769 3-way matching (p/c/n)
7771 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7772 detected as combed (p/c/n + u/b)
7775 The parenthesis at the end indicate the matches that would be used for that
7776 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7779 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7782 Default value is @var{pc_n}.
7785 Mark the main input stream as a pre-processed input, and enable the secondary
7786 input stream as the clean source to pick the fields from. See the filter
7787 introduction for more details. It is similar to the @option{clip2} feature from
7790 Default value is @code{0} (disabled).
7793 Set the field to match from. It is recommended to set this to the same value as
7794 @option{order} unless you experience matching failures with that setting. In
7795 certain circumstances changing the field that is used to match from can have a
7796 large impact on matching performance. Available values are:
7800 Automatic (same value as @option{order}).
7802 Match from the bottom field.
7804 Match from the top field.
7807 Default value is @var{auto}.
7810 Set whether or not chroma is included during the match comparisons. In most
7811 cases it is recommended to leave this enabled. You should set this to @code{0}
7812 only if your clip has bad chroma problems such as heavy rainbowing or other
7813 artifacts. Setting this to @code{0} could also be used to speed things up at
7814 the cost of some accuracy.
7816 Default value is @code{1}.
7820 These define an exclusion band which excludes the lines between @option{y0} and
7821 @option{y1} from being included in the field matching decision. An exclusion
7822 band can be used to ignore subtitles, a logo, or other things that may
7823 interfere with the matching. @option{y0} sets the starting scan line and
7824 @option{y1} sets the ending line; all lines in between @option{y0} and
7825 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7826 @option{y0} and @option{y1} to the same value will disable the feature.
7827 @option{y0} and @option{y1} defaults to @code{0}.
7830 Set the scene change detection threshold as a percentage of maximum change on
7831 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7832 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7833 @option{scthresh} is @code{[0.0, 100.0]}.
7835 Default value is @code{12.0}.
7838 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7839 account the combed scores of matches when deciding what match to use as the
7840 final match. Available values are:
7844 No final matching based on combed scores.
7846 Combed scores are only used when a scene change is detected.
7848 Use combed scores all the time.
7851 Default is @var{sc}.
7854 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7855 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7856 Available values are:
7860 No forced calculation.
7862 Force p/c/n calculations.
7864 Force p/c/n/u/b calculations.
7867 Default value is @var{none}.
7870 This is the area combing threshold used for combed frame detection. This
7871 essentially controls how "strong" or "visible" combing must be to be detected.
7872 Larger values mean combing must be more visible and smaller values mean combing
7873 can be less visible or strong and still be detected. Valid settings are from
7874 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7875 be detected as combed). This is basically a pixel difference value. A good
7876 range is @code{[8, 12]}.
7878 Default value is @code{9}.
7881 Sets whether or not chroma is considered in the combed frame decision. Only
7882 disable this if your source has chroma problems (rainbowing, etc.) that are
7883 causing problems for the combed frame detection with chroma enabled. Actually,
7884 using @option{chroma}=@var{0} is usually more reliable, except for the case
7885 where there is chroma only combing in the source.
7887 Default value is @code{0}.
7891 Respectively set the x-axis and y-axis size of the window used during combed
7892 frame detection. This has to do with the size of the area in which
7893 @option{combpel} pixels are required to be detected as combed for a frame to be
7894 declared combed. See the @option{combpel} parameter description for more info.
7895 Possible values are any number that is a power of 2 starting at 4 and going up
7898 Default value is @code{16}.
7901 The number of combed pixels inside any of the @option{blocky} by
7902 @option{blockx} size blocks on the frame for the frame to be detected as
7903 combed. While @option{cthresh} controls how "visible" the combing must be, this
7904 setting controls "how much" combing there must be in any localized area (a
7905 window defined by the @option{blockx} and @option{blocky} settings) on the
7906 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7907 which point no frames will ever be detected as combed). This setting is known
7908 as @option{MI} in TFM/VFM vocabulary.
7910 Default value is @code{80}.
7913 @anchor{p/c/n/u/b meaning}
7914 @subsection p/c/n/u/b meaning
7916 @subsubsection p/c/n
7918 We assume the following telecined stream:
7921 Top fields: 1 2 2 3 4
7922 Bottom fields: 1 2 3 4 4
7925 The numbers correspond to the progressive frame the fields relate to. Here, the
7926 first two frames are progressive, the 3rd and 4th are combed, and so on.
7928 When @code{fieldmatch} is configured to run a matching from bottom
7929 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7934 B 1 2 3 4 4 <-- matching reference
7943 As a result of the field matching, we can see that some frames get duplicated.
7944 To perform a complete inverse telecine, you need to rely on a decimation filter
7945 after this operation. See for instance the @ref{decimate} filter.
7947 The same operation now matching from top fields (@option{field}=@var{top})
7952 T 1 2 2 3 4 <-- matching reference
7962 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7963 basically, they refer to the frame and field of the opposite parity:
7966 @item @var{p} matches the field of the opposite parity in the previous frame
7967 @item @var{c} matches the field of the opposite parity in the current frame
7968 @item @var{n} matches the field of the opposite parity in the next frame
7973 The @var{u} and @var{b} matching are a bit special in the sense that they match
7974 from the opposite parity flag. In the following examples, we assume that we are
7975 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7976 'x' is placed above and below each matched fields.
7978 With bottom matching (@option{field}=@var{bottom}):
7983 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7984 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7992 With top matching (@option{field}=@var{top}):
7997 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7998 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
8006 @subsection Examples
8008 Simple IVTC of a top field first telecined stream:
8010 fieldmatch=order=tff:combmatch=none, decimate
8013 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
8015 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
8020 Transform the field order of the input video.
8022 It accepts the following parameters:
8027 The output field order. Valid values are @var{tff} for top field first or @var{bff}
8028 for bottom field first.
8031 The default value is @samp{tff}.
8033 The transformation is done by shifting the picture content up or down
8034 by one line, and filling the remaining line with appropriate picture content.
8035 This method is consistent with most broadcast field order converters.
8037 If the input video is not flagged as being interlaced, or it is already
8038 flagged as being of the required output field order, then this filter does
8039 not alter the incoming video.
8041 It is very useful when converting to or from PAL DV material,
8042 which is bottom field first.
8046 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
8049 @section fifo, afifo
8051 Buffer input images and send them when they are requested.
8053 It is mainly useful when auto-inserted by the libavfilter
8056 It does not take parameters.
8060 Find a rectangular object
8062 It accepts the following options:
8066 Filepath of the object image, needs to be in gray8.
8069 Detection threshold, default is 0.5.
8072 Number of mipmaps, default is 3.
8074 @item xmin, ymin, xmax, ymax
8075 Specifies the rectangle in which to search.
8078 @subsection Examples
8082 Generate a representative palette of a given video using @command{ffmpeg}:
8084 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8090 Cover a rectangular object
8092 It accepts the following options:
8096 Filepath of the optional cover image, needs to be in yuv420.
8101 It accepts the following values:
8104 cover it by the supplied image
8106 cover it by interpolating the surrounding pixels
8109 Default value is @var{blur}.
8112 @subsection Examples
8116 Generate a representative palette of a given video using @command{ffmpeg}:
8118 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8125 Convert the input video to one of the specified pixel formats.
8126 Libavfilter will try to pick one that is suitable as input to
8129 It accepts the following parameters:
8133 A '|'-separated list of pixel format names, such as
8134 "pix_fmts=yuv420p|monow|rgb24".
8138 @subsection Examples
8142 Convert the input video to the @var{yuv420p} format
8144 format=pix_fmts=yuv420p
8147 Convert the input video to any of the formats in the list
8149 format=pix_fmts=yuv420p|yuv444p|yuv410p
8156 Convert the video to specified constant frame rate by duplicating or dropping
8157 frames as necessary.
8159 It accepts the following parameters:
8163 The desired output frame rate. The default is @code{25}.
8168 Possible values are:
8171 zero round towards 0
8175 round towards -infinity
8177 round towards +infinity
8181 The default is @code{near}.
8184 Assume the first PTS should be the given value, in seconds. This allows for
8185 padding/trimming at the start of stream. By default, no assumption is made
8186 about the first frame's expected PTS, so no padding or trimming is done.
8187 For example, this could be set to 0 to pad the beginning with duplicates of
8188 the first frame if a video stream starts after the audio stream or to trim any
8189 frames with a negative PTS.
8193 Alternatively, the options can be specified as a flat string:
8194 @var{fps}[:@var{round}].
8196 See also the @ref{setpts} filter.
8198 @subsection Examples
8202 A typical usage in order to set the fps to 25:
8208 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8210 fps=fps=film:round=near
8216 Pack two different video streams into a stereoscopic video, setting proper
8217 metadata on supported codecs. The two views should have the same size and
8218 framerate and processing will stop when the shorter video ends. Please note
8219 that you may conveniently adjust view properties with the @ref{scale} and
8222 It accepts the following parameters:
8226 The desired packing format. Supported values are:
8231 The views are next to each other (default).
8234 The views are on top of each other.
8237 The views are packed by line.
8240 The views are packed by column.
8243 The views are temporally interleaved.
8252 # Convert left and right views into a frame-sequential video
8253 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8255 # Convert views into a side-by-side video with the same output resolution as the input
8256 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
8261 Change the frame rate by interpolating new video output frames from the source
8264 This filter is not designed to function correctly with interlaced media. If
8265 you wish to change the frame rate of interlaced media then you are required
8266 to deinterlace before this filter and re-interlace after this filter.
8268 A description of the accepted options follows.
8272 Specify the output frames per second. This option can also be specified
8273 as a value alone. The default is @code{50}.
8276 Specify the start of a range where the output frame will be created as a
8277 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8278 the default is @code{15}.
8281 Specify the end of a range where the output frame will be created as a
8282 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8283 the default is @code{240}.
8286 Specify the level at which a scene change is detected as a value between
8287 0 and 100 to indicate a new scene; a low value reflects a low
8288 probability for the current frame to introduce a new scene, while a higher
8289 value means the current frame is more likely to be one.
8290 The default is @code{7}.
8293 Specify flags influencing the filter process.
8295 Available value for @var{flags} is:
8298 @item scene_change_detect, scd
8299 Enable scene change detection using the value of the option @var{scene}.
8300 This flag is enabled by default.
8306 Select one frame every N-th frame.
8308 This filter accepts the following option:
8311 Select frame after every @code{step} frames.
8312 Allowed values are positive integers higher than 0. Default value is @code{1}.
8318 Apply a frei0r effect to the input video.
8320 To enable the compilation of this filter, you need to install the frei0r
8321 header and configure FFmpeg with @code{--enable-frei0r}.
8323 It accepts the following parameters:
8328 The name of the frei0r effect to load. If the environment variable
8329 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8330 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8331 Otherwise, the standard frei0r paths are searched, in this order:
8332 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8333 @file{/usr/lib/frei0r-1/}.
8336 A '|'-separated list of parameters to pass to the frei0r effect.
8340 A frei0r effect parameter can be a boolean (its value is either
8341 "y" or "n"), a double, a color (specified as
8342 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8343 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8344 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8345 @var{X} and @var{Y} are floating point numbers) and/or a string.
8347 The number and types of parameters depend on the loaded effect. If an
8348 effect parameter is not specified, the default value is set.
8350 @subsection Examples
8354 Apply the distort0r effect, setting the first two double parameters:
8356 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8360 Apply the colordistance effect, taking a color as the first parameter:
8362 frei0r=colordistance:0.2/0.3/0.4
8363 frei0r=colordistance:violet
8364 frei0r=colordistance:0x112233
8368 Apply the perspective effect, specifying the top left and top right image
8371 frei0r=perspective:0.2/0.2|0.8/0.2
8375 For more information, see
8376 @url{http://frei0r.dyne.org}
8380 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8382 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8383 processing filter, one of them is performed once per block, not per pixel.
8384 This allows for much higher speed.
8386 The filter accepts the following options:
8390 Set quality. This option defines the number of levels for averaging. It accepts
8391 an integer in the range 4-5. Default value is @code{4}.
8394 Force a constant quantization parameter. It accepts an integer in range 0-63.
8395 If not set, the filter will use the QP from the video stream (if available).
8398 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8399 more details but also more artifacts, while higher values make the image smoother
8400 but also blurrier. Default value is @code{0} − PSNR optimal.
8403 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8404 option may cause flicker since the B-Frames have often larger QP. Default is
8405 @code{0} (not enabled).
8411 Apply Gaussian blur filter.
8413 The filter accepts the following options:
8417 Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8420 Set number of steps for Gaussian approximation. Defauls is @code{1}.
8423 Set which planes to filter. By default all planes are filtered.
8426 Set vertical sigma, if negative it will be same as @code{sigma}.
8427 Default is @code{-1}.
8432 The filter accepts the following options:
8436 Set the luminance expression.
8438 Set the chrominance blue expression.
8440 Set the chrominance red expression.
8442 Set the alpha expression.
8444 Set the red expression.
8446 Set the green expression.
8448 Set the blue expression.
8451 The colorspace is selected according to the specified options. If one
8452 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8453 options is specified, the filter will automatically select a YCbCr
8454 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8455 @option{blue_expr} options is specified, it will select an RGB
8458 If one of the chrominance expression is not defined, it falls back on the other
8459 one. If no alpha expression is specified it will evaluate to opaque value.
8460 If none of chrominance expressions are specified, they will evaluate
8461 to the luminance expression.
8463 The expressions can use the following variables and functions:
8467 The sequential number of the filtered frame, starting from @code{0}.
8471 The coordinates of the current sample.
8475 The width and height of the image.
8479 Width and height scale depending on the currently filtered plane. It is the
8480 ratio between the corresponding luma plane number of pixels and the current
8481 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8482 @code{0.5,0.5} for chroma planes.
8485 Time of the current frame, expressed in seconds.
8488 Return the value of the pixel at location (@var{x},@var{y}) of the current
8492 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8496 Return the value of the pixel at location (@var{x},@var{y}) of the
8497 blue-difference chroma plane. Return 0 if there is no such plane.
8500 Return the value of the pixel at location (@var{x},@var{y}) of the
8501 red-difference chroma plane. Return 0 if there is no such plane.
8506 Return the value of the pixel at location (@var{x},@var{y}) of the
8507 red/green/blue component. Return 0 if there is no such component.
8510 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8511 plane. Return 0 if there is no such plane.
8514 For functions, if @var{x} and @var{y} are outside the area, the value will be
8515 automatically clipped to the closer edge.
8517 @subsection Examples
8521 Flip the image horizontally:
8527 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8528 wavelength of 100 pixels:
8530 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8534 Generate a fancy enigmatic moving light:
8536 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
8540 Generate a quick emboss effect:
8542 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8546 Modify RGB components depending on pixel position:
8548 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8552 Create a radial gradient that is the same size as the input (also see
8553 the @ref{vignette} filter):
8555 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8561 Fix the banding artifacts that are sometimes introduced into nearly flat
8562 regions by truncation to 8-bit color depth.
8563 Interpolate the gradients that should go where the bands are, and
8566 It is designed for playback only. Do not use it prior to
8567 lossy compression, because compression tends to lose the dither and
8568 bring back the bands.
8570 It accepts the following parameters:
8575 The maximum amount by which the filter will change any one pixel. This is also
8576 the threshold for detecting nearly flat regions. Acceptable values range from
8577 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8581 The neighborhood to fit the gradient to. A larger radius makes for smoother
8582 gradients, but also prevents the filter from modifying the pixels near detailed
8583 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8584 values will be clipped to the valid range.
8588 Alternatively, the options can be specified as a flat string:
8589 @var{strength}[:@var{radius}]
8591 @subsection Examples
8595 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8601 Specify radius, omitting the strength (which will fall-back to the default
8612 Apply a Hald CLUT to a video stream.
8614 First input is the video stream to process, and second one is the Hald CLUT.
8615 The Hald CLUT input can be a simple picture or a complete video stream.
8617 The filter accepts the following options:
8621 Force termination when the shortest input terminates. Default is @code{0}.
8623 Continue applying the last CLUT after the end of the stream. A value of
8624 @code{0} disable the filter after the last frame of the CLUT is reached.
8625 Default is @code{1}.
8628 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8629 filters share the same internals).
8631 More information about the Hald CLUT can be found on Eskil Steenberg's website
8632 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8634 @subsection Workflow examples
8636 @subsubsection Hald CLUT video stream
8638 Generate an identity Hald CLUT stream altered with various effects:
8640 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
8643 Note: make sure you use a lossless codec.
8645 Then use it with @code{haldclut} to apply it on some random stream:
8647 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8650 The Hald CLUT will be applied to the 10 first seconds (duration of
8651 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8652 to the remaining frames of the @code{mandelbrot} stream.
8654 @subsubsection Hald CLUT with preview
8656 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8657 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8658 biggest possible square starting at the top left of the picture. The remaining
8659 padding pixels (bottom or right) will be ignored. This area can be used to add
8660 a preview of the Hald CLUT.
8662 Typically, the following generated Hald CLUT will be supported by the
8663 @code{haldclut} filter:
8666 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8667 pad=iw+320 [padded_clut];
8668 smptebars=s=320x256, split [a][b];
8669 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8670 [main][b] overlay=W-320" -frames:v 1 clut.png
8673 It contains the original and a preview of the effect of the CLUT: SMPTE color
8674 bars are displayed on the right-top, and below the same color bars processed by
8677 Then, the effect of this Hald CLUT can be visualized with:
8679 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8684 Flip the input video horizontally.
8686 For example, to horizontally flip the input video with @command{ffmpeg}:
8688 ffmpeg -i in.avi -vf "hflip" out.avi
8692 This filter applies a global color histogram equalization on a
8695 It can be used to correct video that has a compressed range of pixel
8696 intensities. The filter redistributes the pixel intensities to
8697 equalize their distribution across the intensity range. It may be
8698 viewed as an "automatically adjusting contrast filter". This filter is
8699 useful only for correcting degraded or poorly captured source
8702 The filter accepts the following options:
8706 Determine the amount of equalization to be applied. As the strength
8707 is reduced, the distribution of pixel intensities more-and-more
8708 approaches that of the input frame. The value must be a float number
8709 in the range [0,1] and defaults to 0.200.
8712 Set the maximum intensity that can generated and scale the output
8713 values appropriately. The strength should be set as desired and then
8714 the intensity can be limited if needed to avoid washing-out. The value
8715 must be a float number in the range [0,1] and defaults to 0.210.
8718 Set the antibanding level. If enabled the filter will randomly vary
8719 the luminance of output pixels by a small amount to avoid banding of
8720 the histogram. Possible values are @code{none}, @code{weak} or
8721 @code{strong}. It defaults to @code{none}.
8726 Compute and draw a color distribution histogram for the input video.
8728 The computed histogram is a representation of the color component
8729 distribution in an image.
8731 Standard histogram displays the color components distribution in an image.
8732 Displays color graph for each color component. Shows distribution of
8733 the Y, U, V, A or R, G, B components, depending on input format, in the
8734 current frame. Below each graph a color component scale meter is shown.
8736 The filter accepts the following options:
8740 Set height of level. Default value is @code{200}.
8741 Allowed range is [50, 2048].
8744 Set height of color scale. Default value is @code{12}.
8745 Allowed range is [0, 40].
8749 It accepts the following values:
8752 Per color component graphs are placed below each other.
8755 Presents information identical to that in the @code{parade}, except
8756 that the graphs representing color components are superimposed directly
8759 Default is @code{parade}.
8762 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8763 Default is @code{linear}.
8766 Set what color components to display.
8767 Default is @code{7}.
8770 Set foreground opacity. Default is @code{0.7}.
8773 Set background opacity. Default is @code{0.5}.
8776 @subsection Examples
8781 Calculate and draw histogram:
8783 ffplay -i input -vf histogram
8791 This is a high precision/quality 3d denoise filter. It aims to reduce
8792 image noise, producing smooth images and making still images really
8793 still. It should enhance compressibility.
8795 It accepts the following optional parameters:
8799 A non-negative floating point number which specifies spatial luma strength.
8802 @item chroma_spatial
8803 A non-negative floating point number which specifies spatial chroma strength.
8804 It defaults to 3.0*@var{luma_spatial}/4.0.
8807 A floating point number which specifies luma temporal strength. It defaults to
8808 6.0*@var{luma_spatial}/4.0.
8811 A floating point number which specifies chroma temporal strength. It defaults to
8812 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8815 @anchor{hwupload_cuda}
8816 @section hwupload_cuda
8818 Upload system memory frames to a CUDA device.
8820 It accepts the following optional parameters:
8824 The number of the CUDA device to use
8829 Apply a high-quality magnification filter designed for pixel art. This filter
8830 was originally created by Maxim Stepin.
8832 It accepts the following option:
8836 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8837 @code{hq3x} and @code{4} for @code{hq4x}.
8838 Default is @code{3}.
8842 Stack input videos horizontally.
8844 All streams must be of same pixel format and of same height.
8846 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8847 to create same output.
8849 The filter accept the following option:
8853 Set number of input streams. Default is 2.
8856 If set to 1, force the output to terminate when the shortest input
8857 terminates. Default value is 0.
8862 Modify the hue and/or the saturation of the input.
8864 It accepts the following parameters:
8868 Specify the hue angle as a number of degrees. It accepts an expression,
8869 and defaults to "0".
8872 Specify the saturation in the [-10,10] range. It accepts an expression and
8876 Specify the hue angle as a number of radians. It accepts an
8877 expression, and defaults to "0".
8880 Specify the brightness in the [-10,10] range. It accepts an expression and
8884 @option{h} and @option{H} are mutually exclusive, and can't be
8885 specified at the same time.
8887 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8888 expressions containing the following constants:
8892 frame count of the input frame starting from 0
8895 presentation timestamp of the input frame expressed in time base units
8898 frame rate of the input video, NAN if the input frame rate is unknown
8901 timestamp expressed in seconds, NAN if the input timestamp is unknown
8904 time base of the input video
8907 @subsection Examples
8911 Set the hue to 90 degrees and the saturation to 1.0:
8917 Same command but expressing the hue in radians:
8923 Rotate hue and make the saturation swing between 0
8924 and 2 over a period of 1 second:
8926 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8930 Apply a 3 seconds saturation fade-in effect starting at 0:
8935 The general fade-in expression can be written as:
8937 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8941 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8943 hue="s=max(0\, min(1\, (8-t)/3))"
8946 The general fade-out expression can be written as:
8948 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8953 @subsection Commands
8955 This filter supports the following commands:
8961 Modify the hue and/or the saturation and/or brightness of the input video.
8962 The command accepts the same syntax of the corresponding option.
8964 If the specified expression is not valid, it is kept at its current
8970 Grow first stream into second stream by connecting components.
8971 This makes it possible to build more robust edge masks.
8973 This filter accepts the following options:
8977 Set which planes will be processed as bitmap, unprocessed planes will be
8978 copied from first stream.
8979 By default value 0xf, all planes will be processed.
8982 Set threshold which is used in filtering. If pixel component value is higher than
8983 this value filter algorithm for connecting components is activated.
8984 By default value is 0.
8989 Detect video interlacing type.
8991 This filter tries to detect if the input frames are interlaced, progressive,
8992 top or bottom field first. It will also try to detect fields that are
8993 repeated between adjacent frames (a sign of telecine).
8995 Single frame detection considers only immediately adjacent frames when classifying each frame.
8996 Multiple frame detection incorporates the classification history of previous frames.
8998 The filter will log these metadata values:
9001 @item single.current_frame
9002 Detected type of current frame using single-frame detection. One of:
9003 ``tff'' (top field first), ``bff'' (bottom field first),
9004 ``progressive'', or ``undetermined''
9007 Cumulative number of frames detected as top field first using single-frame detection.
9010 Cumulative number of frames detected as top field first using multiple-frame detection.
9013 Cumulative number of frames detected as bottom field first using single-frame detection.
9015 @item multiple.current_frame
9016 Detected type of current frame using multiple-frame detection. One of:
9017 ``tff'' (top field first), ``bff'' (bottom field first),
9018 ``progressive'', or ``undetermined''
9021 Cumulative number of frames detected as bottom field first using multiple-frame detection.
9023 @item single.progressive
9024 Cumulative number of frames detected as progressive using single-frame detection.
9026 @item multiple.progressive
9027 Cumulative number of frames detected as progressive using multiple-frame detection.
9029 @item single.undetermined
9030 Cumulative number of frames that could not be classified using single-frame detection.
9032 @item multiple.undetermined
9033 Cumulative number of frames that could not be classified using multiple-frame detection.
9035 @item repeated.current_frame
9036 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
9038 @item repeated.neither
9039 Cumulative number of frames with no repeated field.
9042 Cumulative number of frames with the top field repeated from the previous frame's top field.
9044 @item repeated.bottom
9045 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
9048 The filter accepts the following options:
9052 Set interlacing threshold.
9054 Set progressive threshold.
9056 Threshold for repeated field detection.
9058 Number of frames after which a given frame's contribution to the
9059 statistics is halved (i.e., it contributes only 0.5 to its
9060 classification). The default of 0 means that all frames seen are given
9061 full weight of 1.0 forever.
9062 @item analyze_interlaced_flag
9063 When this is not 0 then idet will use the specified number of frames to determine
9064 if the interlaced flag is accurate, it will not count undetermined frames.
9065 If the flag is found to be accurate it will be used without any further
9066 computations, if it is found to be inaccurate it will be cleared without any
9067 further computations. This allows inserting the idet filter as a low computational
9068 method to clean up the interlaced flag
9073 Deinterleave or interleave fields.
9075 This filter allows one to process interlaced images fields without
9076 deinterlacing them. Deinterleaving splits the input frame into 2
9077 fields (so called half pictures). Odd lines are moved to the top
9078 half of the output image, even lines to the bottom half.
9079 You can process (filter) them independently and then re-interleave them.
9081 The filter accepts the following options:
9085 @item chroma_mode, c
9087 Available values for @var{luma_mode}, @var{chroma_mode} and
9088 @var{alpha_mode} are:
9094 @item deinterleave, d
9095 Deinterleave fields, placing one above the other.
9098 Interleave fields. Reverse the effect of deinterleaving.
9100 Default value is @code{none}.
9103 @item chroma_swap, cs
9104 @item alpha_swap, as
9105 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9110 Apply inflate effect to the video.
9112 This filter replaces the pixel by the local(3x3) average by taking into account
9113 only values higher than the pixel.
9115 It accepts the following options:
9122 Limit the maximum change for each plane, default is 65535.
9123 If 0, plane will remain unchanged.
9128 Simple interlacing filter from progressive contents. This interleaves upper (or
9129 lower) lines from odd frames with lower (or upper) lines from even frames,
9130 halving the frame rate and preserving image height.
9133 Original Original New Frame
9134 Frame 'j' Frame 'j+1' (tff)
9135 ========== =========== ==================
9136 Line 0 --------------------> Frame 'j' Line 0
9137 Line 1 Line 1 ----> Frame 'j+1' Line 1
9138 Line 2 ---------------------> Frame 'j' Line 2
9139 Line 3 Line 3 ----> Frame 'j+1' Line 3
9141 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9144 It accepts the following optional parameters:
9148 This determines whether the interlaced frame is taken from the even
9149 (tff - default) or odd (bff) lines of the progressive frame.
9152 Enable (default) or disable the vertical lowpass filter to avoid twitter
9153 interlacing and reduce moire patterns.
9158 Deinterlace input video by applying Donald Graft's adaptive kernel
9159 deinterling. Work on interlaced parts of a video to produce
9162 The description of the accepted parameters follows.
9166 Set the threshold which affects the filter's tolerance when
9167 determining if a pixel line must be processed. It must be an integer
9168 in the range [0,255] and defaults to 10. A value of 0 will result in
9169 applying the process on every pixels.
9172 Paint pixels exceeding the threshold value to white if set to 1.
9176 Set the fields order. Swap fields if set to 1, leave fields alone if
9180 Enable additional sharpening if set to 1. Default is 0.
9183 Enable twoway sharpening if set to 1. Default is 0.
9186 @subsection Examples
9190 Apply default values:
9192 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9196 Enable additional sharpening:
9202 Paint processed pixels in white:
9208 @section lenscorrection
9210 Correct radial lens distortion
9212 This filter can be used to correct for radial distortion as can result from the use
9213 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9214 one can use tools available for example as part of opencv or simply trial-and-error.
9215 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9216 and extract the k1 and k2 coefficients from the resulting matrix.
9218 Note that effectively the same filter is available in the open-source tools Krita and
9219 Digikam from the KDE project.
9221 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9222 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9223 brightness distribution, so you may want to use both filters together in certain
9224 cases, though you will have to take care of ordering, i.e. whether vignetting should
9225 be applied before or after lens correction.
9229 The filter accepts the following options:
9233 Relative x-coordinate of the focal point of the image, and thereby the center of the
9234 distortion. This value has a range [0,1] and is expressed as fractions of the image
9237 Relative y-coordinate of the focal point of the image, and thereby the center of the
9238 distortion. This value has a range [0,1] and is expressed as fractions of the image
9241 Coefficient of the quadratic correction term. 0.5 means no correction.
9243 Coefficient of the double quadratic correction term. 0.5 means no correction.
9246 The formula that generates the correction is:
9248 @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)
9250 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9251 distances from the focal point in the source and target images, respectively.
9257 The filter accepts the following options:
9261 Set the number of loops.
9264 Set maximal size in number of frames.
9267 Set first frame of loop.
9273 Apply a 3D LUT to an input video.
9275 The filter accepts the following options:
9279 Set the 3D LUT file name.
9281 Currently supported formats:
9293 Select interpolation mode.
9295 Available values are:
9299 Use values from the nearest defined point.
9301 Interpolate values using the 8 points defining a cube.
9303 Interpolate values using a tetrahedron.
9309 Turn certain luma values into transparency.
9311 The filter accepts the following options:
9315 Set the luma which will be used as base for transparency.
9316 Default value is @code{0}.
9319 Set the range of luma values to be keyed out.
9320 Default value is @code{0}.
9323 Set the range of softness. Default value is @code{0}.
9324 Use this to control gradual transition from zero to full transparency.
9327 @section lut, lutrgb, lutyuv
9329 Compute a look-up table for binding each pixel component input value
9330 to an output value, and apply it to the input video.
9332 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9333 to an RGB input video.
9335 These filters accept the following parameters:
9338 set first pixel component expression
9340 set second pixel component expression
9342 set third pixel component expression
9344 set fourth pixel component expression, corresponds to the alpha component
9347 set red component expression
9349 set green component expression
9351 set blue component expression
9353 alpha component expression
9356 set Y/luminance component expression
9358 set U/Cb component expression
9360 set V/Cr component expression
9363 Each of them specifies the expression to use for computing the lookup table for
9364 the corresponding pixel component values.
9366 The exact component associated to each of the @var{c*} options depends on the
9369 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9370 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9372 The expressions can contain the following constants and functions:
9377 The input width and height.
9380 The input value for the pixel component.
9383 The input value, clipped to the @var{minval}-@var{maxval} range.
9386 The maximum value for the pixel component.
9389 The minimum value for the pixel component.
9392 The negated value for the pixel component value, clipped to the
9393 @var{minval}-@var{maxval} range; it corresponds to the expression
9394 "maxval-clipval+minval".
9397 The computed value in @var{val}, clipped to the
9398 @var{minval}-@var{maxval} range.
9400 @item gammaval(gamma)
9401 The computed gamma correction value of the pixel component value,
9402 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9404 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9408 All expressions default to "val".
9410 @subsection Examples
9416 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9417 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9420 The above is the same as:
9422 lutrgb="r=negval:g=negval:b=negval"
9423 lutyuv="y=negval:u=negval:v=negval"
9433 Remove chroma components, turning the video into a graytone image:
9435 lutyuv="u=128:v=128"
9439 Apply a luma burning effect:
9445 Remove green and blue components:
9451 Set a constant alpha channel value on input:
9453 format=rgba,lutrgb=a="maxval-minval/2"
9457 Correct luminance gamma by a factor of 0.5:
9459 lutyuv=y=gammaval(0.5)
9463 Discard least significant bits of luma:
9465 lutyuv=y='bitand(val, 128+64+32)'
9469 Technicolor like effect:
9471 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9477 Compute and apply a lookup table from two video inputs.
9479 This filter accepts the following parameters:
9482 set first pixel component expression
9484 set second pixel component expression
9486 set third pixel component expression
9488 set fourth pixel component expression, corresponds to the alpha component
9491 Each of them specifies the expression to use for computing the lookup table for
9492 the corresponding pixel component values.
9494 The exact component associated to each of the @var{c*} options depends on the
9497 The expressions can contain the following constants:
9502 The input width and height.
9505 The first input value for the pixel component.
9508 The second input value for the pixel component.
9511 The first input video bit depth.
9514 The second input video bit depth.
9517 All expressions default to "x".
9519 @subsection Examples
9523 Highlight differences between two RGB video streams:
9525 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)'
9529 Highlight differences between two YUV video streams:
9531 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)'
9535 @section maskedclamp
9537 Clamp the first input stream with the second input and third input stream.
9539 Returns the value of first stream to be between second input
9540 stream - @code{undershoot} and third input stream + @code{overshoot}.
9542 This filter accepts the following options:
9545 Default value is @code{0}.
9548 Default value is @code{0}.
9551 Set which planes will be processed as bitmap, unprocessed planes will be
9552 copied from first stream.
9553 By default value 0xf, all planes will be processed.
9556 @section maskedmerge
9558 Merge the first input stream with the second input stream using per pixel
9559 weights in the third input stream.
9561 A value of 0 in the third stream pixel component means that pixel component
9562 from first stream is returned unchanged, while maximum value (eg. 255 for
9563 8-bit videos) means that pixel component from second stream is returned
9564 unchanged. Intermediate values define the amount of merging between both
9565 input stream's pixel components.
9567 This filter accepts the following options:
9570 Set which planes will be processed as bitmap, unprocessed planes will be
9571 copied from first stream.
9572 By default value 0xf, all planes will be processed.
9577 Apply motion-compensation deinterlacing.
9579 It needs one field per frame as input and must thus be used together
9580 with yadif=1/3 or equivalent.
9582 This filter accepts the following options:
9585 Set the deinterlacing mode.
9587 It accepts one of the following values:
9592 use iterative motion estimation
9594 like @samp{slow}, but use multiple reference frames.
9596 Default value is @samp{fast}.
9599 Set the picture field parity assumed for the input video. It must be
9600 one of the following values:
9604 assume top field first
9606 assume bottom field first
9609 Default value is @samp{bff}.
9612 Set per-block quantization parameter (QP) used by the internal
9615 Higher values should result in a smoother motion vector field but less
9616 optimal individual vectors. Default value is 1.
9619 @section mergeplanes
9621 Merge color channel components from several video streams.
9623 The filter accepts up to 4 input streams, and merge selected input
9624 planes to the output video.
9626 This filter accepts the following options:
9629 Set input to output plane mapping. Default is @code{0}.
9631 The mappings is specified as a bitmap. It should be specified as a
9632 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9633 mapping for the first plane of the output stream. 'A' sets the number of
9634 the input stream to use (from 0 to 3), and 'a' the plane number of the
9635 corresponding input to use (from 0 to 3). The rest of the mappings is
9636 similar, 'Bb' describes the mapping for the output stream second
9637 plane, 'Cc' describes the mapping for the output stream third plane and
9638 'Dd' describes the mapping for the output stream fourth plane.
9641 Set output pixel format. Default is @code{yuva444p}.
9644 @subsection Examples
9648 Merge three gray video streams of same width and height into single video stream:
9650 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9654 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9656 [a0][a1]mergeplanes=0x00010210:yuva444p
9660 Swap Y and A plane in yuva444p stream:
9662 format=yuva444p,mergeplanes=0x03010200:yuva444p
9666 Swap U and V plane in yuv420p stream:
9668 format=yuv420p,mergeplanes=0x000201:yuv420p
9672 Cast a rgb24 clip to yuv444p:
9674 format=rgb24,mergeplanes=0x000102:yuv444p
9680 Estimate and export motion vectors using block matching algorithms.
9681 Motion vectors are stored in frame side data to be used by other filters.
9683 This filter accepts the following options:
9686 Specify the motion estimation method. Accepts one of the following values:
9690 Exhaustive search algorithm.
9692 Three step search algorithm.
9694 Two dimensional logarithmic search algorithm.
9696 New three step search algorithm.
9698 Four step search algorithm.
9700 Diamond search algorithm.
9702 Hexagon-based search algorithm.
9704 Enhanced predictive zonal search algorithm.
9706 Uneven multi-hexagon search algorithm.
9708 Default value is @samp{esa}.
9711 Macroblock size. Default @code{16}.
9714 Search parameter. Default @code{7}.
9717 @section midequalizer
9719 Apply Midway Image Equalization effect using two video streams.
9721 Midway Image Equalization adjusts a pair of images to have the same
9722 histogram, while maintaining their dynamics as much as possible. It's
9723 useful for e.g. matching exposures from a pair of stereo cameras.
9725 This filter has two inputs and one output, which must be of same pixel format, but
9726 may be of different sizes. The output of filter is first input adjusted with
9727 midway histogram of both inputs.
9729 This filter accepts the following option:
9733 Set which planes to process. Default is @code{15}, which is all available planes.
9736 @section minterpolate
9738 Convert the video to specified frame rate using motion interpolation.
9740 This filter accepts the following options:
9743 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}.
9746 Motion interpolation mode. Following values are accepted:
9749 Duplicate previous or next frame for interpolating new ones.
9751 Blend source frames. Interpolated frame is mean of previous and next frames.
9753 Motion compensated interpolation. Following options are effective when this mode is selected:
9757 Motion compensation mode. Following values are accepted:
9760 Overlapped block motion compensation.
9762 Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9764 Default mode is @samp{obmc}.
9767 Motion estimation mode. Following values are accepted:
9770 Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9772 Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9774 Default mode is @samp{bilat}.
9777 The algorithm to be used for motion estimation. Following values are accepted:
9780 Exhaustive search algorithm.
9782 Three step search algorithm.
9784 Two dimensional logarithmic search algorithm.
9786 New three step search algorithm.
9788 Four step search algorithm.
9790 Diamond search algorithm.
9792 Hexagon-based search algorithm.
9794 Enhanced predictive zonal search algorithm.
9796 Uneven multi-hexagon search algorithm.
9798 Default algorithm is @samp{epzs}.
9801 Macroblock size. Default @code{16}.
9804 Motion estimation search parameter. Default @code{32}.
9807 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).
9812 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:
9815 Disable scene change detection.
9817 Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9819 Default method is @samp{fdiff}.
9822 Scene change detection threshold. Default is @code{5.0}.
9827 Drop frames that do not differ greatly from the previous frame in
9828 order to reduce frame rate.
9830 The main use of this filter is for very-low-bitrate encoding
9831 (e.g. streaming over dialup modem), but it could in theory be used for
9832 fixing movies that were inverse-telecined incorrectly.
9834 A description of the accepted options follows.
9838 Set the maximum number of consecutive frames which can be dropped (if
9839 positive), or the minimum interval between dropped frames (if
9840 negative). If the value is 0, the frame is dropped unregarding the
9841 number of previous sequentially dropped frames.
9848 Set the dropping threshold values.
9850 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9851 represent actual pixel value differences, so a threshold of 64
9852 corresponds to 1 unit of difference for each pixel, or the same spread
9853 out differently over the block.
9855 A frame is a candidate for dropping if no 8x8 blocks differ by more
9856 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9857 meaning the whole image) differ by more than a threshold of @option{lo}.
9859 Default value for @option{hi} is 64*12, default value for @option{lo} is
9860 64*5, and default value for @option{frac} is 0.33.
9868 It accepts an integer in input; if non-zero it negates the
9869 alpha component (if available). The default value in input is 0.
9873 Denoise frames using Non-Local Means algorithm.
9875 Each pixel is adjusted by looking for other pixels with similar contexts. This
9876 context similarity is defined by comparing their surrounding patches of size
9877 @option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9880 Note that the research area defines centers for patches, which means some
9881 patches will be made of pixels outside that research area.
9883 The filter accepts the following options.
9887 Set denoising strength.
9893 Same as @option{p} but for chroma planes.
9895 The default value is @var{0} and means automatic.
9901 Same as @option{r} but for chroma planes.
9903 The default value is @var{0} and means automatic.
9908 Deinterlace video using neural network edge directed interpolation.
9910 This filter accepts the following options:
9914 Mandatory option, without binary file filter can not work.
9915 Currently file can be found here:
9916 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9919 Set which frames to deinterlace, by default it is @code{all}.
9920 Can be @code{all} or @code{interlaced}.
9923 Set mode of operation.
9925 Can be one of the following:
9929 Use frame flags, both fields.
9931 Use frame flags, single field.
9935 Use bottom field only.
9937 Use both fields, top first.
9939 Use both fields, bottom first.
9943 Set which planes to process, by default filter process all frames.
9946 Set size of local neighborhood around each pixel, used by the predictor neural
9949 Can be one of the following:
9962 Set the number of neurons in predicctor neural network.
9963 Can be one of the following:
9974 Controls the number of different neural network predictions that are blended
9975 together to compute the final output value. Can be @code{fast}, default or
9979 Set which set of weights to use in the predictor.
9980 Can be one of the following:
9984 weights trained to minimize absolute error
9986 weights trained to minimize squared error
9990 Controls whether or not the prescreener neural network is used to decide
9991 which pixels should be processed by the predictor neural network and which
9992 can be handled by simple cubic interpolation.
9993 The prescreener is trained to know whether cubic interpolation will be
9994 sufficient for a pixel or whether it should be predicted by the predictor nn.
9995 The computational complexity of the prescreener nn is much less than that of
9996 the predictor nn. Since most pixels can be handled by cubic interpolation,
9997 using the prescreener generally results in much faster processing.
9998 The prescreener is pretty accurate, so the difference between using it and not
9999 using it is almost always unnoticeable.
10001 Can be one of the following:
10009 Default is @code{new}.
10012 Set various debugging flags.
10017 Force libavfilter not to use any of the specified pixel formats for the
10018 input to the next filter.
10020 It accepts the following parameters:
10024 A '|'-separated list of pixel format names, such as
10025 apix_fmts=yuv420p|monow|rgb24".
10029 @subsection Examples
10033 Force libavfilter to use a format different from @var{yuv420p} for the
10034 input to the vflip filter:
10036 noformat=pix_fmts=yuv420p,vflip
10040 Convert the input video to any of the formats not contained in the list:
10042 noformat=yuv420p|yuv444p|yuv410p
10048 Add noise on video input frame.
10050 The filter accepts the following options:
10058 Set noise seed for specific pixel component or all pixel components in case
10059 of @var{all_seed}. Default value is @code{123457}.
10061 @item all_strength, alls
10062 @item c0_strength, c0s
10063 @item c1_strength, c1s
10064 @item c2_strength, c2s
10065 @item c3_strength, c3s
10066 Set noise strength for specific pixel component or all pixel components in case
10067 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
10069 @item all_flags, allf
10070 @item c0_flags, c0f
10071 @item c1_flags, c1f
10072 @item c2_flags, c2f
10073 @item c3_flags, c3f
10074 Set pixel component flags or set flags for all components if @var{all_flags}.
10075 Available values for component flags are:
10078 averaged temporal noise (smoother)
10080 mix random noise with a (semi)regular pattern
10082 temporal noise (noise pattern changes between frames)
10084 uniform noise (gaussian otherwise)
10088 @subsection Examples
10090 Add temporal and uniform noise to input video:
10092 noise=alls=20:allf=t+u
10097 Pass the video source unchanged to the output.
10100 Optical Character Recognition
10102 This filter uses Tesseract for optical character recognition.
10104 It accepts the following options:
10108 Set datapath to tesseract data. Default is to use whatever was
10109 set at installation.
10112 Set language, default is "eng".
10115 Set character whitelist.
10118 Set character blacklist.
10121 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10125 Apply a video transform using libopencv.
10127 To enable this filter, install the libopencv library and headers and
10128 configure FFmpeg with @code{--enable-libopencv}.
10130 It accepts the following parameters:
10135 The name of the libopencv filter to apply.
10137 @item filter_params
10138 The parameters to pass to the libopencv filter. If not specified, the default
10139 values are assumed.
10143 Refer to the official libopencv documentation for more precise
10145 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10147 Several libopencv filters are supported; see the following subsections.
10152 Dilate an image by using a specific structuring element.
10153 It corresponds to the libopencv function @code{cvDilate}.
10155 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10157 @var{struct_el} represents a structuring element, and has the syntax:
10158 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10160 @var{cols} and @var{rows} represent the number of columns and rows of
10161 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10162 point, and @var{shape} the shape for the structuring element. @var{shape}
10163 must be "rect", "cross", "ellipse", or "custom".
10165 If the value for @var{shape} is "custom", it must be followed by a
10166 string of the form "=@var{filename}". The file with name
10167 @var{filename} is assumed to represent a binary image, with each
10168 printable character corresponding to a bright pixel. When a custom
10169 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10170 or columns and rows of the read file are assumed instead.
10172 The default value for @var{struct_el} is "3x3+0x0/rect".
10174 @var{nb_iterations} specifies the number of times the transform is
10175 applied to the image, and defaults to 1.
10179 # Use the default values
10182 # Dilate using a structuring element with a 5x5 cross, iterating two times
10183 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10185 # Read the shape from the file diamond.shape, iterating two times.
10186 # The file diamond.shape may contain a pattern of characters like this
10192 # The specified columns and rows are ignored
10193 # but the anchor point coordinates are not
10194 ocv=dilate:0x0+2x2/custom=diamond.shape|2
10199 Erode an image by using a specific structuring element.
10200 It corresponds to the libopencv function @code{cvErode}.
10202 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10203 with the same syntax and semantics as the @ref{dilate} filter.
10207 Smooth the input video.
10209 The filter takes the following parameters:
10210 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10212 @var{type} is the type of smooth filter to apply, and must be one of
10213 the following values: "blur", "blur_no_scale", "median", "gaussian",
10214 or "bilateral". The default value is "gaussian".
10216 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10217 depend on the smooth type. @var{param1} and
10218 @var{param2} accept integer positive values or 0. @var{param3} and
10219 @var{param4} accept floating point values.
10221 The default value for @var{param1} is 3. The default value for the
10222 other parameters is 0.
10224 These parameters correspond to the parameters assigned to the
10225 libopencv function @code{cvSmooth}.
10230 Overlay one video on top of another.
10232 It takes two inputs and has one output. The first input is the "main"
10233 video on which the second input is overlaid.
10235 It accepts the following parameters:
10237 A description of the accepted options follows.
10242 Set the expression for the x and y coordinates of the overlaid video
10243 on the main video. Default value is "0" for both expressions. In case
10244 the expression is invalid, it is set to a huge value (meaning that the
10245 overlay will not be displayed within the output visible area).
10248 The action to take when EOF is encountered on the secondary input; it accepts
10249 one of the following values:
10253 Repeat the last frame (the default).
10257 Pass the main input through.
10261 Set when the expressions for @option{x}, and @option{y} are evaluated.
10263 It accepts the following values:
10266 only evaluate expressions once during the filter initialization or
10267 when a command is processed
10270 evaluate expressions for each incoming frame
10273 Default value is @samp{frame}.
10276 If set to 1, force the output to terminate when the shortest input
10277 terminates. Default value is 0.
10280 Set the format for the output video.
10282 It accepts the following values:
10285 force YUV420 output
10288 force YUV422 output
10291 force YUV444 output
10294 force packed RGB output
10297 force planar RGB output
10300 Default value is @samp{yuv420}.
10302 @item rgb @emph{(deprecated)}
10303 If set to 1, force the filter to accept inputs in the RGB
10304 color space. Default value is 0. This option is deprecated, use
10305 @option{format} instead.
10308 If set to 1, force the filter to draw the last overlay frame over the
10309 main input until the end of the stream. A value of 0 disables this
10310 behavior. Default value is 1.
10313 The @option{x}, and @option{y} expressions can contain the following
10319 The main input width and height.
10323 The overlay input width and height.
10327 The computed values for @var{x} and @var{y}. They are evaluated for
10332 horizontal and vertical chroma subsample values of the output
10333 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10337 the number of input frame, starting from 0
10340 the position in the file of the input frame, NAN if unknown
10343 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10347 Note that the @var{n}, @var{pos}, @var{t} variables are available only
10348 when evaluation is done @emph{per frame}, and will evaluate to NAN
10349 when @option{eval} is set to @samp{init}.
10351 Be aware that frames are taken from each input video in timestamp
10352 order, hence, if their initial timestamps differ, it is a good idea
10353 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10354 have them begin in the same zero timestamp, as the example for
10355 the @var{movie} filter does.
10357 You can chain together more overlays but you should test the
10358 efficiency of such approach.
10360 @subsection Commands
10362 This filter supports the following commands:
10366 Modify the x and y of the overlay input.
10367 The command accepts the same syntax of the corresponding option.
10369 If the specified expression is not valid, it is kept at its current
10373 @subsection Examples
10377 Draw the overlay at 10 pixels from the bottom right corner of the main
10380 overlay=main_w-overlay_w-10:main_h-overlay_h-10
10383 Using named options the example above becomes:
10385 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10389 Insert a transparent PNG logo in the bottom left corner of the input,
10390 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10392 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10396 Insert 2 different transparent PNG logos (second logo on bottom
10397 right corner) using the @command{ffmpeg} tool:
10399 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
10403 Add a transparent color layer on top of the main video; @code{WxH}
10404 must specify the size of the main input to the overlay filter:
10406 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10410 Play an original video and a filtered version (here with the deshake
10411 filter) side by side using the @command{ffplay} tool:
10413 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10416 The above command is the same as:
10418 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10422 Make a sliding overlay appearing from the left to the right top part of the
10423 screen starting since time 2:
10425 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10429 Compose output by putting two input videos side to side:
10431 ffmpeg -i left.avi -i right.avi -filter_complex "
10432 nullsrc=size=200x100 [background];
10433 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10434 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10435 [background][left] overlay=shortest=1 [background+left];
10436 [background+left][right] overlay=shortest=1:x=100 [left+right]
10441 Mask 10-20 seconds of a video by applying the delogo filter to a section
10443 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10444 -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]'
10449 Chain several overlays in cascade:
10451 nullsrc=s=200x200 [bg];
10452 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10453 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10454 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10455 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10456 [in3] null, [mid2] overlay=100:100 [out0]
10463 Apply Overcomplete Wavelet denoiser.
10465 The filter accepts the following options:
10471 Larger depth values will denoise lower frequency components more, but
10472 slow down filtering.
10474 Must be an int in the range 8-16, default is @code{8}.
10476 @item luma_strength, ls
10479 Must be a double value in the range 0-1000, default is @code{1.0}.
10481 @item chroma_strength, cs
10482 Set chroma strength.
10484 Must be a double value in the range 0-1000, default is @code{1.0}.
10490 Add paddings to the input image, and place the original input at the
10491 provided @var{x}, @var{y} coordinates.
10493 It accepts the following parameters:
10498 Specify an expression for the size of the output image with the
10499 paddings added. If the value for @var{width} or @var{height} is 0, the
10500 corresponding input size is used for the output.
10502 The @var{width} expression can reference the value set by the
10503 @var{height} expression, and vice versa.
10505 The default value of @var{width} and @var{height} is 0.
10509 Specify the offsets to place the input image at within the padded area,
10510 with respect to the top/left border of the output image.
10512 The @var{x} expression can reference the value set by the @var{y}
10513 expression, and vice versa.
10515 The default value of @var{x} and @var{y} is 0.
10517 If @var{x} or @var{y} evaluate to a negative number, they'll be changed
10518 so the input image is centered on the padded area.
10521 Specify the color of the padded area. For the syntax of this option,
10522 check the "Color" section in the ffmpeg-utils manual.
10524 The default value of @var{color} is "black".
10527 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10529 It accepts the following values:
10533 Only evaluate expressions once during the filter initialization or when
10534 a command is processed.
10537 Evaluate expressions for each incoming frame.
10541 Default value is @samp{init}.
10544 Pad to aspect instead to a resolution.
10548 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10549 options are expressions containing the following constants:
10554 The input video width and height.
10558 These are the same as @var{in_w} and @var{in_h}.
10562 The output width and height (the size of the padded area), as
10563 specified by the @var{width} and @var{height} expressions.
10567 These are the same as @var{out_w} and @var{out_h}.
10571 The x and y offsets as specified by the @var{x} and @var{y}
10572 expressions, or NAN if not yet specified.
10575 same as @var{iw} / @var{ih}
10578 input sample aspect ratio
10581 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10585 The horizontal and vertical chroma subsample values. For example for the
10586 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10589 @subsection Examples
10593 Add paddings with the color "violet" to the input video. The output video
10594 size is 640x480, and the top-left corner of the input video is placed at
10597 pad=640:480:0:40:violet
10600 The example above is equivalent to the following command:
10602 pad=width=640:height=480:x=0:y=40:color=violet
10606 Pad the input to get an output with dimensions increased by 3/2,
10607 and put the input video at the center of the padded area:
10609 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10613 Pad the input to get a squared output with size equal to the maximum
10614 value between the input width and height, and put the input video at
10615 the center of the padded area:
10617 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10621 Pad the input to get a final w/h ratio of 16:9:
10623 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10627 In case of anamorphic video, in order to set the output display aspect
10628 correctly, it is necessary to use @var{sar} in the expression,
10629 according to the relation:
10631 (ih * X / ih) * sar = output_dar
10632 X = output_dar / sar
10635 Thus the previous example needs to be modified to:
10637 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10641 Double the output size and put the input video in the bottom-right
10642 corner of the output padded area:
10644 pad="2*iw:2*ih:ow-iw:oh-ih"
10648 @anchor{palettegen}
10649 @section palettegen
10651 Generate one palette for a whole video stream.
10653 It accepts the following options:
10657 Set the maximum number of colors to quantize in the palette.
10658 Note: the palette will still contain 256 colors; the unused palette entries
10661 @item reserve_transparent
10662 Create a palette of 255 colors maximum and reserve the last one for
10663 transparency. Reserving the transparency color is useful for GIF optimization.
10664 If not set, the maximum of colors in the palette will be 256. You probably want
10665 to disable this option for a standalone image.
10669 Set statistics mode.
10671 It accepts the following values:
10674 Compute full frame histograms.
10676 Compute histograms only for the part that differs from previous frame. This
10677 might be relevant to give more importance to the moving part of your input if
10678 the background is static.
10680 Compute new histogram for each frame.
10683 Default value is @var{full}.
10686 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10687 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10688 color quantization of the palette. This information is also visible at
10689 @var{info} logging level.
10691 @subsection Examples
10695 Generate a representative palette of a given video using @command{ffmpeg}:
10697 ffmpeg -i input.mkv -vf palettegen palette.png
10701 @section paletteuse
10703 Use a palette to downsample an input video stream.
10705 The filter takes two inputs: one video stream and a palette. The palette must
10706 be a 256 pixels image.
10708 It accepts the following options:
10712 Select dithering mode. Available algorithms are:
10715 Ordered 8x8 bayer dithering (deterministic)
10717 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10718 Note: this dithering is sometimes considered "wrong" and is included as a
10720 @item floyd_steinberg
10721 Floyd and Steingberg dithering (error diffusion)
10723 Frankie Sierra dithering v2 (error diffusion)
10725 Frankie Sierra dithering v2 "Lite" (error diffusion)
10728 Default is @var{sierra2_4a}.
10731 When @var{bayer} dithering is selected, this option defines the scale of the
10732 pattern (how much the crosshatch pattern is visible). A low value means more
10733 visible pattern for less banding, and higher value means less visible pattern
10734 at the cost of more banding.
10736 The option must be an integer value in the range [0,5]. Default is @var{2}.
10739 If set, define the zone to process
10743 Only the changing rectangle will be reprocessed. This is similar to GIF
10744 cropping/offsetting compression mechanism. This option can be useful for speed
10745 if only a part of the image is changing, and has use cases such as limiting the
10746 scope of the error diffusal @option{dither} to the rectangle that bounds the
10747 moving scene (it leads to more deterministic output if the scene doesn't change
10748 much, and as a result less moving noise and better GIF compression).
10751 Default is @var{none}.
10754 Take new palette for each output frame.
10757 @subsection Examples
10761 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10762 using @command{ffmpeg}:
10764 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10768 @section perspective
10770 Correct perspective of video not recorded perpendicular to the screen.
10772 A description of the accepted parameters follows.
10783 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10784 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10785 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10786 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10787 then the corners of the source will be sent to the specified coordinates.
10789 The expressions can use the following variables:
10794 the width and height of video frame.
10798 Output frame count.
10801 @item interpolation
10802 Set interpolation for perspective correction.
10804 It accepts the following values:
10810 Default value is @samp{linear}.
10813 Set interpretation of coordinate options.
10815 It accepts the following values:
10819 Send point in the source specified by the given coordinates to
10820 the corners of the destination.
10822 @item 1, destination
10824 Send the corners of the source to the point in the destination specified
10825 by the given coordinates.
10827 Default value is @samp{source}.
10831 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10833 It accepts the following values:
10836 only evaluate expressions once during the filter initialization or
10837 when a command is processed
10840 evaluate expressions for each incoming frame
10843 Default value is @samp{init}.
10848 Delay interlaced video by one field time so that the field order changes.
10850 The intended use is to fix PAL movies that have been captured with the
10851 opposite field order to the film-to-video transfer.
10853 A description of the accepted parameters follows.
10859 It accepts the following values:
10862 Capture field order top-first, transfer bottom-first.
10863 Filter will delay the bottom field.
10866 Capture field order bottom-first, transfer top-first.
10867 Filter will delay the top field.
10870 Capture and transfer with the same field order. This mode only exists
10871 for the documentation of the other options to refer to, but if you
10872 actually select it, the filter will faithfully do nothing.
10875 Capture field order determined automatically by field flags, transfer
10877 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10878 basis using field flags. If no field information is available,
10879 then this works just like @samp{u}.
10882 Capture unknown or varying, transfer opposite.
10883 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10884 analyzing the images and selecting the alternative that produces best
10885 match between the fields.
10888 Capture top-first, transfer unknown or varying.
10889 Filter selects among @samp{t} and @samp{p} using image analysis.
10892 Capture bottom-first, transfer unknown or varying.
10893 Filter selects among @samp{b} and @samp{p} using image analysis.
10896 Capture determined by field flags, transfer unknown or varying.
10897 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10898 image analysis. If no field information is available, then this works just
10899 like @samp{U}. This is the default mode.
10902 Both capture and transfer unknown or varying.
10903 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10907 @section pixdesctest
10909 Pixel format descriptor test filter, mainly useful for internal
10910 testing. The output video should be equal to the input video.
10914 format=monow, pixdesctest
10917 can be used to test the monowhite pixel format descriptor definition.
10921 Enable the specified chain of postprocessing subfilters using libpostproc. This
10922 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10923 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10924 Each subfilter and some options have a short and a long name that can be used
10925 interchangeably, i.e. dr/dering are the same.
10927 The filters accept the following options:
10931 Set postprocessing subfilters string.
10934 All subfilters share common options to determine their scope:
10938 Honor the quality commands for this subfilter.
10941 Do chrominance filtering, too (default).
10944 Do luminance filtering only (no chrominance).
10947 Do chrominance filtering only (no luminance).
10950 These options can be appended after the subfilter name, separated by a '|'.
10952 Available subfilters are:
10955 @item hb/hdeblock[|difference[|flatness]]
10956 Horizontal deblocking filter
10959 Difference factor where higher values mean more deblocking (default: @code{32}).
10961 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10964 @item vb/vdeblock[|difference[|flatness]]
10965 Vertical deblocking filter
10968 Difference factor where higher values mean more deblocking (default: @code{32}).
10970 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10973 @item ha/hadeblock[|difference[|flatness]]
10974 Accurate horizontal deblocking filter
10977 Difference factor where higher values mean more deblocking (default: @code{32}).
10979 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10982 @item va/vadeblock[|difference[|flatness]]
10983 Accurate vertical deblocking filter
10986 Difference factor where higher values mean more deblocking (default: @code{32}).
10988 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10992 The horizontal and vertical deblocking filters share the difference and
10993 flatness values so you cannot set different horizontal and vertical
10997 @item h1/x1hdeblock
10998 Experimental horizontal deblocking filter
11000 @item v1/x1vdeblock
11001 Experimental vertical deblocking filter
11006 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
11009 larger -> stronger filtering
11011 larger -> stronger filtering
11013 larger -> stronger filtering
11016 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
11019 Stretch luminance to @code{0-255}.
11022 @item lb/linblenddeint
11023 Linear blend deinterlacing filter that deinterlaces the given block by
11024 filtering all lines with a @code{(1 2 1)} filter.
11026 @item li/linipoldeint
11027 Linear interpolating deinterlacing filter that deinterlaces the given block by
11028 linearly interpolating every second line.
11030 @item ci/cubicipoldeint
11031 Cubic interpolating deinterlacing filter deinterlaces the given block by
11032 cubically interpolating every second line.
11034 @item md/mediandeint
11035 Median deinterlacing filter that deinterlaces the given block by applying a
11036 median filter to every second line.
11038 @item fd/ffmpegdeint
11039 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
11040 second line with a @code{(-1 4 2 4 -1)} filter.
11043 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
11044 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
11046 @item fq/forceQuant[|quantizer]
11047 Overrides the quantizer table from the input with the constant quantizer you
11055 Default pp filter combination (@code{hb|a,vb|a,dr|a})
11058 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
11061 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
11064 @subsection Examples
11068 Apply horizontal and vertical deblocking, deringing and automatic
11069 brightness/contrast:
11075 Apply default filters without brightness/contrast correction:
11081 Apply default filters and temporal denoiser:
11083 pp=default/tmpnoise|1|2|3
11087 Apply deblocking on luminance only, and switch vertical deblocking on or off
11088 automatically depending on available CPU time:
11095 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11096 similar to spp = 6 with 7 point DCT, where only the center sample is
11099 The filter accepts the following options:
11103 Force a constant quantization parameter. It accepts an integer in range
11104 0 to 63. If not set, the filter will use the QP from the video stream
11108 Set thresholding mode. Available modes are:
11112 Set hard thresholding.
11114 Set soft thresholding (better de-ringing effect, but likely blurrier).
11116 Set medium thresholding (good results, default).
11120 @section premultiply
11121 Apply alpha premultiply effect to input video stream using first plane
11122 of second stream as alpha.
11124 Both streams must have same dimensions and same pixel format.
11126 The filter accepts the following option:
11130 Set which planes will be processed, unprocessed planes will be copied.
11131 By default value 0xf, all planes will be processed.
11135 Apply prewitt operator to input video stream.
11137 The filter accepts the following option:
11141 Set which planes will be processed, unprocessed planes will be copied.
11142 By default value 0xf, all planes will be processed.
11145 Set value which will be multiplied with filtered result.
11148 Set value which will be added to filtered result.
11153 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11154 Ratio) between two input videos.
11156 This filter takes in input two input videos, the first input is
11157 considered the "main" source and is passed unchanged to the
11158 output. The second input is used as a "reference" video for computing
11161 Both video inputs must have the same resolution and pixel format for
11162 this filter to work correctly. Also it assumes that both inputs
11163 have the same number of frames, which are compared one by one.
11165 The obtained average PSNR is printed through the logging system.
11167 The filter stores the accumulated MSE (mean squared error) of each
11168 frame, and at the end of the processing it is averaged across all frames
11169 equally, and the following formula is applied to obtain the PSNR:
11172 PSNR = 10*log10(MAX^2/MSE)
11175 Where MAX is the average of the maximum values of each component of the
11178 The description of the accepted parameters follows.
11181 @item stats_file, f
11182 If specified the filter will use the named file to save the PSNR of
11183 each individual frame. When filename equals "-" the data is sent to
11186 @item stats_version
11187 Specifies which version of the stats file format to use. Details of
11188 each format are written below.
11189 Default value is 1.
11191 @item stats_add_max
11192 Determines whether the max value is output to the stats log.
11193 Default value is 0.
11194 Requires stats_version >= 2. If this is set and stats_version < 2,
11195 the filter will return an error.
11198 The file printed if @var{stats_file} is selected, contains a sequence of
11199 key/value pairs of the form @var{key}:@var{value} for each compared
11202 If a @var{stats_version} greater than 1 is specified, a header line precedes
11203 the list of per-frame-pair stats, with key value pairs following the frame
11204 format with the following parameters:
11207 @item psnr_log_version
11208 The version of the log file format. Will match @var{stats_version}.
11211 A comma separated list of the per-frame-pair parameters included in
11215 A description of each shown per-frame-pair parameter follows:
11219 sequential number of the input frame, starting from 1
11222 Mean Square Error pixel-by-pixel average difference of the compared
11223 frames, averaged over all the image components.
11225 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11226 Mean Square Error pixel-by-pixel average difference of the compared
11227 frames for the component specified by the suffix.
11229 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11230 Peak Signal to Noise ratio of the compared frames for the component
11231 specified by the suffix.
11233 @item max_avg, max_y, max_u, max_v
11234 Maximum allowed value for each channel, and average over all
11240 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11241 [main][ref] psnr="stats_file=stats.log" [out]
11244 On this example the input file being processed is compared with the
11245 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11246 is stored in @file{stats.log}.
11251 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11252 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11255 The pullup filter is designed to take advantage of future context in making
11256 its decisions. This filter is stateless in the sense that it does not lock
11257 onto a pattern to follow, but it instead looks forward to the following
11258 fields in order to identify matches and rebuild progressive frames.
11260 To produce content with an even framerate, insert the fps filter after
11261 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11262 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11264 The filter accepts the following options:
11271 These options set the amount of "junk" to ignore at the left, right, top, and
11272 bottom of the image, respectively. Left and right are in units of 8 pixels,
11273 while top and bottom are in units of 2 lines.
11274 The default is 8 pixels on each side.
11277 Set the strict breaks. Setting this option to 1 will reduce the chances of
11278 filter generating an occasional mismatched frame, but it may also cause an
11279 excessive number of frames to be dropped during high motion sequences.
11280 Conversely, setting it to -1 will make filter match fields more easily.
11281 This may help processing of video where there is slight blurring between
11282 the fields, but may also cause there to be interlaced frames in the output.
11283 Default value is @code{0}.
11286 Set the metric plane to use. It accepts the following values:
11292 Use chroma blue plane.
11295 Use chroma red plane.
11298 This option may be set to use chroma plane instead of the default luma plane
11299 for doing filter's computations. This may improve accuracy on very clean
11300 source material, but more likely will decrease accuracy, especially if there
11301 is chroma noise (rainbow effect) or any grayscale video.
11302 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11303 load and make pullup usable in realtime on slow machines.
11306 For best results (without duplicated frames in the output file) it is
11307 necessary to change the output frame rate. For example, to inverse
11308 telecine NTSC input:
11310 ffmpeg -i input -vf pullup -r 24000/1001 ...
11315 Change video quantization parameters (QP).
11317 The filter accepts the following option:
11321 Set expression for quantization parameter.
11324 The expression is evaluated through the eval API and can contain, among others,
11325 the following constants:
11329 1 if index is not 129, 0 otherwise.
11332 Sequentional index starting from -129 to 128.
11335 @subsection Examples
11339 Some equation like:
11347 Flush video frames from internal cache of frames into a random order.
11348 No frame is discarded.
11349 Inspired by @ref{frei0r} nervous filter.
11353 Set size in number of frames of internal cache, in range from @code{2} to
11354 @code{512}. Default is @code{30}.
11357 Set seed for random number generator, must be an integer included between
11358 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11359 less than @code{0}, the filter will try to use a good random seed on a
11363 @section readeia608
11365 Read closed captioning (EIA-608) information from the top lines of a video frame.
11367 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11368 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11369 with EIA-608 data (starting from 0). A description of each metadata value follows:
11372 @item lavfi.readeia608.X.cc
11373 The two bytes stored as EIA-608 data (printed in hexadecimal).
11375 @item lavfi.readeia608.X.line
11376 The number of the line on which the EIA-608 data was identified and read.
11379 This filter accepts the following options:
11383 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11386 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11389 Set minimal acceptable amplitude change for sync codes detection.
11390 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11393 Set the ratio of width reserved for sync code detection.
11394 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11397 Set the max peaks height difference for sync code detection.
11398 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11401 Set max peaks period difference for sync code detection.
11402 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11405 Set the first two max start code bits differences.
11406 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11409 Set the minimum ratio of bits height compared to 3rd start code bit.
11410 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11413 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11416 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11419 Enable checking the parity bit. In the event of a parity error, the filter will output
11420 @code{0x00} for that character. Default is false.
11423 @subsection Examples
11427 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11429 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
11435 Read vertical interval timecode (VITC) information from the top lines of a
11438 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11439 timecode value, if a valid timecode has been detected. Further metadata key
11440 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11441 timecode data has been found or not.
11443 This filter accepts the following options:
11447 Set the maximum number of lines to scan for VITC data. If the value is set to
11448 @code{-1} the full video frame is scanned. Default is @code{45}.
11451 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11452 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11455 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11456 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11459 @subsection Examples
11463 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11464 draw @code{--:--:--:--} as a placeholder:
11466 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11472 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11474 Destination pixel at position (X, Y) will be picked from source (x, y) position
11475 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11476 value for pixel will be used for destination pixel.
11478 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11479 will have Xmap/Ymap video stream dimensions.
11480 Xmap and Ymap input video streams are 16bit depth, single channel.
11482 @section removegrain
11484 The removegrain filter is a spatial denoiser for progressive video.
11488 Set mode for the first plane.
11491 Set mode for the second plane.
11494 Set mode for the third plane.
11497 Set mode for the fourth plane.
11500 Range of mode is from 0 to 24. Description of each mode follows:
11504 Leave input plane unchanged. Default.
11507 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11510 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11513 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11516 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11517 This is equivalent to a median filter.
11520 Line-sensitive clipping giving the minimal change.
11523 Line-sensitive clipping, intermediate.
11526 Line-sensitive clipping, intermediate.
11529 Line-sensitive clipping, intermediate.
11532 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11535 Replaces the target pixel with the closest neighbour.
11538 [1 2 1] horizontal and vertical kernel blur.
11544 Bob mode, interpolates top field from the line where the neighbours
11545 pixels are the closest.
11548 Bob mode, interpolates bottom field from the line where the neighbours
11549 pixels are the closest.
11552 Bob mode, interpolates top field. Same as 13 but with a more complicated
11553 interpolation formula.
11556 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11557 interpolation formula.
11560 Clips the pixel with the minimum and maximum of respectively the maximum and
11561 minimum of each pair of opposite neighbour pixels.
11564 Line-sensitive clipping using opposite neighbours whose greatest distance from
11565 the current pixel is minimal.
11568 Replaces the pixel with the average of its 8 neighbours.
11571 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11574 Clips pixels using the averages of opposite neighbour.
11577 Same as mode 21 but simpler and faster.
11580 Small edge and halo removal, but reputed useless.
11586 @section removelogo
11588 Suppress a TV station logo, using an image file to determine which
11589 pixels comprise the logo. It works by filling in the pixels that
11590 comprise the logo with neighboring pixels.
11592 The filter accepts the following options:
11596 Set the filter bitmap file, which can be any image format supported by
11597 libavformat. The width and height of the image file must match those of the
11598 video stream being processed.
11601 Pixels in the provided bitmap image with a value of zero are not
11602 considered part of the logo, non-zero pixels are considered part of
11603 the logo. If you use white (255) for the logo and black (0) for the
11604 rest, you will be safe. For making the filter bitmap, it is
11605 recommended to take a screen capture of a black frame with the logo
11606 visible, and then using a threshold filter followed by the erode
11607 filter once or twice.
11609 If needed, little splotches can be fixed manually. Remember that if
11610 logo pixels are not covered, the filter quality will be much
11611 reduced. Marking too many pixels as part of the logo does not hurt as
11612 much, but it will increase the amount of blurring needed to cover over
11613 the image and will destroy more information than necessary, and extra
11614 pixels will slow things down on a large logo.
11616 @section repeatfields
11618 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11619 fields based on its value.
11623 Reverse a video clip.
11625 Warning: This filter requires memory to buffer the entire clip, so trimming
11628 @subsection Examples
11632 Take the first 5 seconds of a clip, and reverse it.
11640 Rotate video by an arbitrary angle expressed in radians.
11642 The filter accepts the following options:
11644 A description of the optional parameters follows.
11647 Set an expression for the angle by which to rotate the input video
11648 clockwise, expressed as a number of radians. A negative value will
11649 result in a counter-clockwise rotation. By default it is set to "0".
11651 This expression is evaluated for each frame.
11654 Set the output width expression, default value is "iw".
11655 This expression is evaluated just once during configuration.
11658 Set the output height expression, default value is "ih".
11659 This expression is evaluated just once during configuration.
11662 Enable bilinear interpolation if set to 1, a value of 0 disables
11663 it. Default value is 1.
11666 Set the color used to fill the output area not covered by the rotated
11667 image. For the general syntax of this option, check the "Color" section in the
11668 ffmpeg-utils manual. If the special value "none" is selected then no
11669 background is printed (useful for example if the background is never shown).
11671 Default value is "black".
11674 The expressions for the angle and the output size can contain the
11675 following constants and functions:
11679 sequential number of the input frame, starting from 0. It is always NAN
11680 before the first frame is filtered.
11683 time in seconds of the input frame, it is set to 0 when the filter is
11684 configured. It is always NAN before the first frame is filtered.
11688 horizontal and vertical chroma subsample values. For example for the
11689 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11693 the input video width and height
11697 the output width and height, that is the size of the padded area as
11698 specified by the @var{width} and @var{height} expressions
11702 the minimal width/height required for completely containing the input
11703 video rotated by @var{a} radians.
11705 These are only available when computing the @option{out_w} and
11706 @option{out_h} expressions.
11709 @subsection Examples
11713 Rotate the input by PI/6 radians clockwise:
11719 Rotate the input by PI/6 radians counter-clockwise:
11725 Rotate the input by 45 degrees clockwise:
11731 Apply a constant rotation with period T, starting from an angle of PI/3:
11733 rotate=PI/3+2*PI*t/T
11737 Make the input video rotation oscillating with a period of T
11738 seconds and an amplitude of A radians:
11740 rotate=A*sin(2*PI/T*t)
11744 Rotate the video, output size is chosen so that the whole rotating
11745 input video is always completely contained in the output:
11747 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11751 Rotate the video, reduce the output size so that no background is ever
11754 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11758 @subsection Commands
11760 The filter supports the following commands:
11764 Set the angle expression.
11765 The command accepts the same syntax of the corresponding option.
11767 If the specified expression is not valid, it is kept at its current
11773 Apply Shape Adaptive Blur.
11775 The filter accepts the following options:
11778 @item luma_radius, lr
11779 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11780 value is 1.0. A greater value will result in a more blurred image, and
11781 in slower processing.
11783 @item luma_pre_filter_radius, lpfr
11784 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11787 @item luma_strength, ls
11788 Set luma maximum difference between pixels to still be considered, must
11789 be a value in the 0.1-100.0 range, default value is 1.0.
11791 @item chroma_radius, cr
11792 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11793 greater value will result in a more blurred image, and in slower
11796 @item chroma_pre_filter_radius, cpfr
11797 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11799 @item chroma_strength, cs
11800 Set chroma maximum difference between pixels to still be considered,
11801 must be a value in the -0.9-100.0 range.
11804 Each chroma option value, if not explicitly specified, is set to the
11805 corresponding luma option value.
11810 Scale (resize) the input video, using the libswscale library.
11812 The scale filter forces the output display aspect ratio to be the same
11813 of the input, by changing the output sample aspect ratio.
11815 If the input image format is different from the format requested by
11816 the next filter, the scale filter will convert the input to the
11819 @subsection Options
11820 The filter accepts the following options, or any of the options
11821 supported by the libswscale scaler.
11823 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11824 the complete list of scaler options.
11829 Set the output video dimension expression. Default value is the input
11832 If the value is 0, the input width is used for the output.
11834 If one of the values is -1, the scale filter will use a value that
11835 maintains the aspect ratio of the input image, calculated from the
11836 other specified dimension. If both of them are -1, the input size is
11839 If one of the values is -n with n > 1, the scale filter will also use a value
11840 that maintains the aspect ratio of the input image, calculated from the other
11841 specified dimension. After that it will, however, make sure that the calculated
11842 dimension is divisible by n and adjust the value if necessary.
11844 See below for the list of accepted constants for use in the dimension
11848 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11852 Only evaluate expressions once during the filter initialization or when a command is processed.
11855 Evaluate expressions for each incoming frame.
11859 Default value is @samp{init}.
11863 Set the interlacing mode. It accepts the following values:
11867 Force interlaced aware scaling.
11870 Do not apply interlaced scaling.
11873 Select interlaced aware scaling depending on whether the source frames
11874 are flagged as interlaced or not.
11877 Default value is @samp{0}.
11880 Set libswscale scaling flags. See
11881 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11882 complete list of values. If not explicitly specified the filter applies
11886 @item param0, param1
11887 Set libswscale input parameters for scaling algorithms that need them. See
11888 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11889 complete documentation. If not explicitly specified the filter applies
11895 Set the video size. For the syntax of this option, check the
11896 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11898 @item in_color_matrix
11899 @item out_color_matrix
11900 Set in/output YCbCr color space type.
11902 This allows the autodetected value to be overridden as well as allows forcing
11903 a specific value used for the output and encoder.
11905 If not specified, the color space type depends on the pixel format.
11911 Choose automatically.
11914 Format conforming to International Telecommunication Union (ITU)
11915 Recommendation BT.709.
11918 Set color space conforming to the United States Federal Communications
11919 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11922 Set color space conforming to:
11926 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11929 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11932 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11937 Set color space conforming to SMPTE ST 240:1999.
11942 Set in/output YCbCr sample range.
11944 This allows the autodetected value to be overridden as well as allows forcing
11945 a specific value used for the output and encoder. If not specified, the
11946 range depends on the pixel format. Possible values:
11950 Choose automatically.
11953 Set full range (0-255 in case of 8-bit luma).
11956 Set "MPEG" range (16-235 in case of 8-bit luma).
11959 @item force_original_aspect_ratio
11960 Enable decreasing or increasing output video width or height if necessary to
11961 keep the original aspect ratio. Possible values:
11965 Scale the video as specified and disable this feature.
11968 The output video dimensions will automatically be decreased if needed.
11971 The output video dimensions will automatically be increased if needed.
11975 One useful instance of this option is that when you know a specific device's
11976 maximum allowed resolution, you can use this to limit the output video to
11977 that, while retaining the aspect ratio. For example, device A allows
11978 1280x720 playback, and your video is 1920x800. Using this option (set it to
11979 decrease) and specifying 1280x720 to the command line makes the output
11982 Please note that this is a different thing than specifying -1 for @option{w}
11983 or @option{h}, you still need to specify the output resolution for this option
11988 The values of the @option{w} and @option{h} options are expressions
11989 containing the following constants:
11994 The input width and height
11998 These are the same as @var{in_w} and @var{in_h}.
12002 The output (scaled) width and height
12006 These are the same as @var{out_w} and @var{out_h}
12009 The same as @var{iw} / @var{ih}
12012 input sample aspect ratio
12015 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
12019 horizontal and vertical input chroma subsample values. For example for the
12020 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12024 horizontal and vertical output chroma subsample values. For example for the
12025 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12028 @subsection Examples
12032 Scale the input video to a size of 200x100
12037 This is equivalent to:
12048 Specify a size abbreviation for the output size:
12053 which can also be written as:
12059 Scale the input to 2x:
12061 scale=w=2*iw:h=2*ih
12065 The above is the same as:
12067 scale=2*in_w:2*in_h
12071 Scale the input to 2x with forced interlaced scaling:
12073 scale=2*iw:2*ih:interl=1
12077 Scale the input to half size:
12079 scale=w=iw/2:h=ih/2
12083 Increase the width, and set the height to the same size:
12089 Seek Greek harmony:
12096 Increase the height, and set the width to 3/2 of the height:
12098 scale=w=3/2*oh:h=3/5*ih
12102 Increase the size, making the size a multiple of the chroma
12105 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12109 Increase the width to a maximum of 500 pixels,
12110 keeping the same aspect ratio as the input:
12112 scale=w='min(500\, iw*3/2):h=-1'
12116 @subsection Commands
12118 This filter supports the following commands:
12122 Set the output video dimension expression.
12123 The command accepts the same syntax of the corresponding option.
12125 If the specified expression is not valid, it is kept at its current
12131 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12132 format conversion on CUDA video frames. Setting the output width and height
12133 works in the same way as for the @var{scale} filter.
12135 The following additional options are accepted:
12138 The pixel format of the output CUDA frames. If set to the string "same" (the
12139 default), the input format will be kept. Note that automatic format negotiation
12140 and conversion is not yet supported for hardware frames
12143 The interpolation algorithm used for resizing. One of the following:
12150 @item cubic2p_bspline
12151 2-parameter cubic (B=1, C=0)
12153 @item cubic2p_catmullrom
12154 2-parameter cubic (B=0, C=1/2)
12156 @item cubic2p_b05c03
12157 2-parameter cubic (B=1/2, C=3/10)
12169 Scale (resize) the input video, based on a reference video.
12171 See the scale filter for available options, scale2ref supports the same but
12172 uses the reference video instead of the main input as basis.
12174 @subsection Examples
12178 Scale a subtitle stream to match the main video in size before overlaying
12180 'scale2ref[b][a];[a][b]overlay'
12184 @anchor{selectivecolor}
12185 @section selectivecolor
12187 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12188 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12189 by the "purity" of the color (that is, how saturated it already is).
12191 This filter is similar to the Adobe Photoshop Selective Color tool.
12193 The filter accepts the following options:
12196 @item correction_method
12197 Select color correction method.
12199 Available values are:
12202 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12205 Specified adjustments are relative to the original component value.
12207 Default is @code{absolute}.
12209 Adjustments for red pixels (pixels where the red component is the maximum)
12211 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12213 Adjustments for green pixels (pixels where the green component is the maximum)
12215 Adjustments for cyan pixels (pixels where the red component is the minimum)
12217 Adjustments for blue pixels (pixels where the blue component is the maximum)
12219 Adjustments for magenta pixels (pixels where the green component is the minimum)
12221 Adjustments for white pixels (pixels where all components are greater than 128)
12223 Adjustments for all pixels except pure black and pure white
12225 Adjustments for black pixels (pixels where all components are lesser than 128)
12227 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12230 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12231 4 space separated floating point adjustment values in the [-1,1] range,
12232 respectively to adjust the amount of cyan, magenta, yellow and black for the
12233 pixels of its range.
12235 @subsection Examples
12239 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12240 increase magenta by 27% in blue areas:
12242 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12246 Use a Photoshop selective color preset:
12248 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12252 @anchor{separatefields}
12253 @section separatefields
12255 The @code{separatefields} takes a frame-based video input and splits
12256 each frame into its components fields, producing a new half height clip
12257 with twice the frame rate and twice the frame count.
12259 This filter use field-dominance information in frame to decide which
12260 of each pair of fields to place first in the output.
12261 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12263 @section setdar, setsar
12265 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12268 This is done by changing the specified Sample (aka Pixel) Aspect
12269 Ratio, according to the following equation:
12271 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12274 Keep in mind that the @code{setdar} filter does not modify the pixel
12275 dimensions of the video frame. Also, the display aspect ratio set by
12276 this filter may be changed by later filters in the filterchain,
12277 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12280 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12281 the filter output video.
12283 Note that as a consequence of the application of this filter, the
12284 output display aspect ratio will change according to the equation
12287 Keep in mind that the sample aspect ratio set by the @code{setsar}
12288 filter may be changed by later filters in the filterchain, e.g. if
12289 another "setsar" or a "setdar" filter is applied.
12291 It accepts the following parameters:
12294 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12295 Set the aspect ratio used by the filter.
12297 The parameter can be a floating point number string, an expression, or
12298 a string of the form @var{num}:@var{den}, where @var{num} and
12299 @var{den} are the numerator and denominator of the aspect ratio. If
12300 the parameter is not specified, it is assumed the value "0".
12301 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12305 Set the maximum integer value to use for expressing numerator and
12306 denominator when reducing the expressed aspect ratio to a rational.
12307 Default value is @code{100}.
12311 The parameter @var{sar} is an expression containing
12312 the following constants:
12316 These are approximated values for the mathematical constants e
12317 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12320 The input width and height.
12323 These are the same as @var{w} / @var{h}.
12326 The input sample aspect ratio.
12329 The input display aspect ratio. It is the same as
12330 (@var{w} / @var{h}) * @var{sar}.
12333 Horizontal and vertical chroma subsample values. For example, for the
12334 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12337 @subsection Examples
12342 To change the display aspect ratio to 16:9, specify one of the following:
12349 To change the sample aspect ratio to 10:11, specify:
12355 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12356 1000 in the aspect ratio reduction, use the command:
12358 setdar=ratio=16/9:max=1000
12366 Force field for the output video frame.
12368 The @code{setfield} filter marks the interlace type field for the
12369 output frames. It does not change the input frame, but only sets the
12370 corresponding property, which affects how the frame is treated by
12371 following filters (e.g. @code{fieldorder} or @code{yadif}).
12373 The filter accepts the following options:
12378 Available values are:
12382 Keep the same field property.
12385 Mark the frame as bottom-field-first.
12388 Mark the frame as top-field-first.
12391 Mark the frame as progressive.
12397 Show a line containing various information for each input video frame.
12398 The input video is not modified.
12400 The shown line contains a sequence of key/value pairs of the form
12401 @var{key}:@var{value}.
12403 The following values are shown in the output:
12407 The (sequential) number of the input frame, starting from 0.
12410 The Presentation TimeStamp of the input frame, expressed as a number of
12411 time base units. The time base unit depends on the filter input pad.
12414 The Presentation TimeStamp of the input frame, expressed as a number of
12418 The position of the frame in the input stream, or -1 if this information is
12419 unavailable and/or meaningless (for example in case of synthetic video).
12422 The pixel format name.
12425 The sample aspect ratio of the input frame, expressed in the form
12426 @var{num}/@var{den}.
12429 The size of the input frame. For the syntax of this option, check the
12430 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12433 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12434 for bottom field first).
12437 This is 1 if the frame is a key frame, 0 otherwise.
12440 The picture type of the input frame ("I" for an I-frame, "P" for a
12441 P-frame, "B" for a B-frame, or "?" for an unknown type).
12442 Also refer to the documentation of the @code{AVPictureType} enum and of
12443 the @code{av_get_picture_type_char} function defined in
12444 @file{libavutil/avutil.h}.
12447 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12449 @item plane_checksum
12450 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12451 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12454 @section showpalette
12456 Displays the 256 colors palette of each frame. This filter is only relevant for
12457 @var{pal8} pixel format frames.
12459 It accepts the following option:
12463 Set the size of the box used to represent one palette color entry. Default is
12464 @code{30} (for a @code{30x30} pixel box).
12467 @section shuffleframes
12469 Reorder and/or duplicate and/or drop video frames.
12471 It accepts the following parameters:
12475 Set the destination indexes of input frames.
12476 This is space or '|' separated list of indexes that maps input frames to output
12477 frames. Number of indexes also sets maximal value that each index may have.
12478 '-1' index have special meaning and that is to drop frame.
12481 The first frame has the index 0. The default is to keep the input unchanged.
12483 @subsection Examples
12487 Swap second and third frame of every three frames of the input:
12489 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12493 Swap 10th and 1st frame of every ten frames of the input:
12495 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12499 @section shuffleplanes
12501 Reorder and/or duplicate video planes.
12503 It accepts the following parameters:
12508 The index of the input plane to be used as the first output plane.
12511 The index of the input plane to be used as the second output plane.
12514 The index of the input plane to be used as the third output plane.
12517 The index of the input plane to be used as the fourth output plane.
12521 The first plane has the index 0. The default is to keep the input unchanged.
12523 @subsection Examples
12527 Swap the second and third planes of the input:
12529 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12533 @anchor{signalstats}
12534 @section signalstats
12535 Evaluate various visual metrics that assist in determining issues associated
12536 with the digitization of analog video media.
12538 By default the filter will log these metadata values:
12542 Display the minimal Y value contained within the input frame. Expressed in
12546 Display the Y value at the 10% percentile within the input frame. Expressed in
12550 Display the average Y value within the input frame. Expressed in range of
12554 Display the Y value at the 90% percentile within the input frame. Expressed in
12558 Display the maximum Y value contained within the input frame. Expressed in
12562 Display the minimal U value contained within the input frame. Expressed in
12566 Display the U value at the 10% percentile within the input frame. Expressed in
12570 Display the average U value within the input frame. Expressed in range of
12574 Display the U value at the 90% percentile within the input frame. Expressed in
12578 Display the maximum U value contained within the input frame. Expressed in
12582 Display the minimal V value contained within the input frame. Expressed in
12586 Display the V value at the 10% percentile within the input frame. Expressed in
12590 Display the average V value within the input frame. Expressed in range of
12594 Display the V value at the 90% percentile within the input frame. Expressed in
12598 Display the maximum V value contained within the input frame. Expressed in
12602 Display the minimal saturation value contained within the input frame.
12603 Expressed in range of [0-~181.02].
12606 Display the saturation value at the 10% percentile within the input frame.
12607 Expressed in range of [0-~181.02].
12610 Display the average saturation value within the input frame. Expressed in range
12614 Display the saturation value at the 90% percentile within the input frame.
12615 Expressed in range of [0-~181.02].
12618 Display the maximum saturation value contained within the input frame.
12619 Expressed in range of [0-~181.02].
12622 Display the median value for hue within the input frame. Expressed in range of
12626 Display the average value for hue within the input frame. Expressed in range of
12630 Display the average of sample value difference between all values of the Y
12631 plane in the current frame and corresponding values of the previous input frame.
12632 Expressed in range of [0-255].
12635 Display the average of sample value difference between all values of the U
12636 plane in the current frame and corresponding values of the previous input frame.
12637 Expressed in range of [0-255].
12640 Display the average of sample value difference between all values of the V
12641 plane in the current frame and corresponding values of the previous input frame.
12642 Expressed in range of [0-255].
12645 Display bit depth of Y plane in current frame.
12646 Expressed in range of [0-16].
12649 Display bit depth of U plane in current frame.
12650 Expressed in range of [0-16].
12653 Display bit depth of V plane in current frame.
12654 Expressed in range of [0-16].
12657 The filter accepts the following options:
12663 @option{stat} specify an additional form of image analysis.
12664 @option{out} output video with the specified type of pixel highlighted.
12666 Both options accept the following values:
12670 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12671 unlike the neighboring pixels of the same field. Examples of temporal outliers
12672 include the results of video dropouts, head clogs, or tape tracking issues.
12675 Identify @var{vertical line repetition}. Vertical line repetition includes
12676 similar rows of pixels within a frame. In born-digital video vertical line
12677 repetition is common, but this pattern is uncommon in video digitized from an
12678 analog source. When it occurs in video that results from the digitization of an
12679 analog source it can indicate concealment from a dropout compensator.
12682 Identify pixels that fall outside of legal broadcast range.
12686 Set the highlight color for the @option{out} option. The default color is
12690 @subsection Examples
12694 Output data of various video metrics:
12696 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12700 Output specific data about the minimum and maximum values of the Y plane per frame:
12702 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12706 Playback video while highlighting pixels that are outside of broadcast range in red.
12708 ffplay example.mov -vf signalstats="out=brng:color=red"
12712 Playback video with signalstats metadata drawn over the frame.
12714 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12717 The contents of signalstat_drawtext.txt used in the command are:
12720 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12721 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12722 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12723 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12731 Calculates the MPEG-7 Video Signature. The filter can handle more than one
12732 input. In this case the matching between the inputs can be calculated additionally.
12733 The filter always passes through the first input. The signature of each stream can
12734 be written into a file.
12736 It accepts the following options:
12740 Enable or disable the matching process.
12742 Available values are:
12746 Disable the calculation of a matching (default).
12748 Calculate the matching for the whole video and output whether the whole video
12749 matches or only parts.
12751 Calculate only until a matching is found or the video ends. Should be faster in
12756 Set the number of inputs. The option value must be a non negative integer.
12757 Default value is 1.
12760 Set the path to which the output is written. If there is more than one input,
12761 the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12762 integer), that will be replaced with the input number. If no filename is
12763 specified, no output will be written. This is the default.
12766 Choose the output format.
12768 Available values are:
12772 Use the specified binary representation (default).
12774 Use the specified xml representation.
12778 Set threshold to detect one word as similar. The option value must be an integer
12779 greater than zero. The default value is 9000.
12782 Set threshold to detect all words as similar. The option value must be an integer
12783 greater than zero. The default value is 60000.
12786 Set threshold to detect frames as similar. The option value must be an integer
12787 greater than zero. The default value is 116.
12790 Set the minimum length of a sequence in frames to recognize it as matching
12791 sequence. The option value must be a non negative integer value.
12792 The default value is 0.
12795 Set the minimum relation, that matching frames to all frames must have.
12796 The option value must be a double value between 0 and 1. The default value is 0.5.
12799 @subsection Examples
12803 To calculate the signature of an input video and store it in signature.bin:
12805 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12809 To detect whether two videos match and store the signatures in XML format in
12810 signature0.xml and signature1.xml:
12812 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 -
12820 Blur the input video without impacting the outlines.
12822 It accepts the following options:
12825 @item luma_radius, lr
12826 Set the luma radius. The option value must be a float number in
12827 the range [0.1,5.0] that specifies the variance of the gaussian filter
12828 used to blur the image (slower if larger). Default value is 1.0.
12830 @item luma_strength, ls
12831 Set the luma strength. The option value must be a float number
12832 in the range [-1.0,1.0] that configures the blurring. A value included
12833 in [0.0,1.0] will blur the image whereas a value included in
12834 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12836 @item luma_threshold, lt
12837 Set the luma threshold used as a coefficient to determine
12838 whether a pixel should be blurred or not. The option value must be an
12839 integer in the range [-30,30]. A value of 0 will filter all the image,
12840 a value included in [0,30] will filter flat areas and a value included
12841 in [-30,0] will filter edges. Default value is 0.
12843 @item chroma_radius, cr
12844 Set the chroma radius. The option value must be a float number in
12845 the range [0.1,5.0] that specifies the variance of the gaussian filter
12846 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12848 @item chroma_strength, cs
12849 Set the chroma strength. The option value must be a float number
12850 in the range [-1.0,1.0] that configures the blurring. A value included
12851 in [0.0,1.0] will blur the image whereas a value included in
12852 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12854 @item chroma_threshold, ct
12855 Set the chroma threshold used as a coefficient to determine
12856 whether a pixel should be blurred or not. The option value must be an
12857 integer in the range [-30,30]. A value of 0 will filter all the image,
12858 a value included in [0,30] will filter flat areas and a value included
12859 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12862 If a chroma option is not explicitly set, the corresponding luma value
12867 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12869 This filter takes in input two input videos, the first input is
12870 considered the "main" source and is passed unchanged to the
12871 output. The second input is used as a "reference" video for computing
12874 Both video inputs must have the same resolution and pixel format for
12875 this filter to work correctly. Also it assumes that both inputs
12876 have the same number of frames, which are compared one by one.
12878 The filter stores the calculated SSIM of each frame.
12880 The description of the accepted parameters follows.
12883 @item stats_file, f
12884 If specified the filter will use the named file to save the SSIM of
12885 each individual frame. When filename equals "-" the data is sent to
12889 The file printed if @var{stats_file} is selected, contains a sequence of
12890 key/value pairs of the form @var{key}:@var{value} for each compared
12893 A description of each shown parameter follows:
12897 sequential number of the input frame, starting from 1
12899 @item Y, U, V, R, G, B
12900 SSIM of the compared frames for the component specified by the suffix.
12903 SSIM of the compared frames for the whole frame.
12906 Same as above but in dB representation.
12911 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12912 [main][ref] ssim="stats_file=stats.log" [out]
12915 On this example the input file being processed is compared with the
12916 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12917 is stored in @file{stats.log}.
12919 Another example with both psnr and ssim at same time:
12921 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12926 Convert between different stereoscopic image formats.
12928 The filters accept the following options:
12932 Set stereoscopic image format of input.
12934 Available values for input image formats are:
12937 side by side parallel (left eye left, right eye right)
12940 side by side crosseye (right eye left, left eye right)
12943 side by side parallel with half width resolution
12944 (left eye left, right eye right)
12947 side by side crosseye with half width resolution
12948 (right eye left, left eye right)
12951 above-below (left eye above, right eye below)
12954 above-below (right eye above, left eye below)
12957 above-below with half height resolution
12958 (left eye above, right eye below)
12961 above-below with half height resolution
12962 (right eye above, left eye below)
12965 alternating frames (left eye first, right eye second)
12968 alternating frames (right eye first, left eye second)
12971 interleaved rows (left eye has top row, right eye starts on next row)
12974 interleaved rows (right eye has top row, left eye starts on next row)
12977 interleaved columns, left eye first
12980 interleaved columns, right eye first
12982 Default value is @samp{sbsl}.
12986 Set stereoscopic image format of output.
12990 side by side parallel (left eye left, right eye right)
12993 side by side crosseye (right eye left, left eye right)
12996 side by side parallel with half width resolution
12997 (left eye left, right eye right)
13000 side by side crosseye with half width resolution
13001 (right eye left, left eye right)
13004 above-below (left eye above, right eye below)
13007 above-below (right eye above, left eye below)
13010 above-below with half height resolution
13011 (left eye above, right eye below)
13014 above-below with half height resolution
13015 (right eye above, left eye below)
13018 alternating frames (left eye first, right eye second)
13021 alternating frames (right eye first, left eye second)
13024 interleaved rows (left eye has top row, right eye starts on next row)
13027 interleaved rows (right eye has top row, left eye starts on next row)
13030 anaglyph red/blue gray
13031 (red filter on left eye, blue filter on right eye)
13034 anaglyph red/green gray
13035 (red filter on left eye, green filter on right eye)
13038 anaglyph red/cyan gray
13039 (red filter on left eye, cyan filter on right eye)
13042 anaglyph red/cyan half colored
13043 (red filter on left eye, cyan filter on right eye)
13046 anaglyph red/cyan color
13047 (red filter on left eye, cyan filter on right eye)
13050 anaglyph red/cyan color optimized with the least squares projection of dubois
13051 (red filter on left eye, cyan filter on right eye)
13054 anaglyph green/magenta gray
13055 (green filter on left eye, magenta filter on right eye)
13058 anaglyph green/magenta half colored
13059 (green filter on left eye, magenta filter on right eye)
13062 anaglyph green/magenta colored
13063 (green filter on left eye, magenta filter on right eye)
13066 anaglyph green/magenta color optimized with the least squares projection of dubois
13067 (green filter on left eye, magenta filter on right eye)
13070 anaglyph yellow/blue gray
13071 (yellow filter on left eye, blue filter on right eye)
13074 anaglyph yellow/blue half colored
13075 (yellow filter on left eye, blue filter on right eye)
13078 anaglyph yellow/blue colored
13079 (yellow filter on left eye, blue filter on right eye)
13082 anaglyph yellow/blue color optimized with the least squares projection of dubois
13083 (yellow filter on left eye, blue filter on right eye)
13086 mono output (left eye only)
13089 mono output (right eye only)
13092 checkerboard, left eye first
13095 checkerboard, right eye first
13098 interleaved columns, left eye first
13101 interleaved columns, right eye first
13107 Default value is @samp{arcd}.
13110 @subsection Examples
13114 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13120 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13126 @section streamselect, astreamselect
13127 Select video or audio streams.
13129 The filter accepts the following options:
13133 Set number of inputs. Default is 2.
13136 Set input indexes to remap to outputs.
13139 @subsection Commands
13141 The @code{streamselect} and @code{astreamselect} filter supports the following
13146 Set input indexes to remap to outputs.
13149 @subsection Examples
13153 Select first 5 seconds 1st stream and rest of time 2nd stream:
13155 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13159 Same as above, but for audio:
13161 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13166 Apply sobel operator to input video stream.
13168 The filter accepts the following option:
13172 Set which planes will be processed, unprocessed planes will be copied.
13173 By default value 0xf, all planes will be processed.
13176 Set value which will be multiplied with filtered result.
13179 Set value which will be added to filtered result.
13185 Apply a simple postprocessing filter that compresses and decompresses the image
13186 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13187 and average the results.
13189 The filter accepts the following options:
13193 Set quality. This option defines the number of levels for averaging. It accepts
13194 an integer in the range 0-6. If set to @code{0}, the filter will have no
13195 effect. A value of @code{6} means the higher quality. For each increment of
13196 that value the speed drops by a factor of approximately 2. Default value is
13200 Force a constant quantization parameter. If not set, the filter will use the QP
13201 from the video stream (if available).
13204 Set thresholding mode. Available modes are:
13208 Set hard thresholding (default).
13210 Set soft thresholding (better de-ringing effect, but likely blurrier).
13213 @item use_bframe_qp
13214 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13215 option may cause flicker since the B-Frames have often larger QP. Default is
13216 @code{0} (not enabled).
13222 Draw subtitles on top of input video using the libass library.
13224 To enable compilation of this filter you need to configure FFmpeg with
13225 @code{--enable-libass}. This filter also requires a build with libavcodec and
13226 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13227 Alpha) subtitles format.
13229 The filter accepts the following options:
13233 Set the filename of the subtitle file to read. It must be specified.
13235 @item original_size
13236 Specify the size of the original video, the video for which the ASS file
13237 was composed. For the syntax of this option, check the
13238 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13239 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13240 correctly scale the fonts if the aspect ratio has been changed.
13243 Set a directory path containing fonts that can be used by the filter.
13244 These fonts will be used in addition to whatever the font provider uses.
13247 Set subtitles input character encoding. @code{subtitles} filter only. Only
13248 useful if not UTF-8.
13250 @item stream_index, si
13251 Set subtitles stream index. @code{subtitles} filter only.
13254 Override default style or script info parameters of the subtitles. It accepts a
13255 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13258 If the first key is not specified, it is assumed that the first value
13259 specifies the @option{filename}.
13261 For example, to render the file @file{sub.srt} on top of the input
13262 video, use the command:
13267 which is equivalent to:
13269 subtitles=filename=sub.srt
13272 To render the default subtitles stream from file @file{video.mkv}, use:
13274 subtitles=video.mkv
13277 To render the second subtitles stream from that file, use:
13279 subtitles=video.mkv:si=1
13282 To make the subtitles stream from @file{sub.srt} appear in transparent green
13283 @code{DejaVu Serif}, use:
13285 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13288 @section super2xsai
13290 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13291 Interpolate) pixel art scaling algorithm.
13293 Useful for enlarging pixel art images without reducing sharpness.
13297 Swap two rectangular objects in video.
13299 This filter accepts the following options:
13309 Set 1st rect x coordinate.
13312 Set 1st rect y coordinate.
13315 Set 2nd rect x coordinate.
13318 Set 2nd rect y coordinate.
13320 All expressions are evaluated once for each frame.
13323 The all options are expressions containing the following constants:
13328 The input width and height.
13331 same as @var{w} / @var{h}
13334 input sample aspect ratio
13337 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13340 The number of the input frame, starting from 0.
13343 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13346 the position in the file of the input frame, NAN if unknown
13354 Apply telecine process to the video.
13356 This filter accepts the following options:
13365 The default value is @code{top}.
13369 A string of numbers representing the pulldown pattern you wish to apply.
13370 The default value is @code{23}.
13374 Some typical patterns:
13379 24p: 2332 (preferred)
13386 24p: 222222222223 ("Euro pulldown")
13393 Apply threshold effect to video stream.
13395 This filter needs four video streams to perform thresholding.
13396 First stream is stream we are filtering.
13397 Second stream is holding threshold values, third stream is holding min values,
13398 and last, fourth stream is holding max values.
13400 The filter accepts the following option:
13404 Set which planes will be processed, unprocessed planes will be copied.
13405 By default value 0xf, all planes will be processed.
13408 For example if first stream pixel's component value is less then threshold value
13409 of pixel component from 2nd threshold stream, third stream value will picked,
13410 otherwise fourth stream pixel component value will be picked.
13412 Using color source filter one can perform various types of thresholding:
13414 @subsection Examples
13418 Binary threshold, using gray color as threshold:
13420 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13424 Inverted binary threshold, using gray color as threshold:
13426 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13430 Truncate binary threshold, using gray color as threshold:
13432 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13436 Threshold to zero, using gray color as threshold:
13438 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13442 Inverted threshold to zero, using gray color as threshold:
13444 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13449 Select the most representative frame in a given sequence of consecutive frames.
13451 The filter accepts the following options:
13455 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13456 will pick one of them, and then handle the next batch of @var{n} frames until
13457 the end. Default is @code{100}.
13460 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13461 value will result in a higher memory usage, so a high value is not recommended.
13463 @subsection Examples
13467 Extract one picture each 50 frames:
13473 Complete example of a thumbnail creation with @command{ffmpeg}:
13475 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13481 Tile several successive frames together.
13483 The filter accepts the following options:
13488 Set the grid size (i.e. the number of lines and columns). For the syntax of
13489 this option, check the
13490 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13493 Set the maximum number of frames to render in the given area. It must be less
13494 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13495 the area will be used.
13498 Set the outer border margin in pixels.
13501 Set the inner border thickness (i.e. the number of pixels between frames). For
13502 more advanced padding options (such as having different values for the edges),
13503 refer to the pad video filter.
13506 Specify the color of the unused area. For the syntax of this option, check the
13507 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13511 @subsection Examples
13515 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13517 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13519 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13520 duplicating each output frame to accommodate the originally detected frame
13524 Display @code{5} pictures in an area of @code{3x2} frames,
13525 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13526 mixed flat and named options:
13528 tile=3x2:nb_frames=5:padding=7:margin=2
13532 @section tinterlace
13534 Perform various types of temporal field interlacing.
13536 Frames are counted starting from 1, so the first input frame is
13539 The filter accepts the following options:
13544 Specify the mode of the interlacing. This option can also be specified
13545 as a value alone. See below for a list of values for this option.
13547 Available values are:
13551 Move odd frames into the upper field, even into the lower field,
13552 generating a double height frame at half frame rate.
13556 Frame 1 Frame 2 Frame 3 Frame 4
13558 11111 22222 33333 44444
13559 11111 22222 33333 44444
13560 11111 22222 33333 44444
13561 11111 22222 33333 44444
13575 Only output odd frames, even frames are dropped, generating a frame with
13576 unchanged height at half frame rate.
13581 Frame 1 Frame 2 Frame 3 Frame 4
13583 11111 22222 33333 44444
13584 11111 22222 33333 44444
13585 11111 22222 33333 44444
13586 11111 22222 33333 44444
13596 Only output even frames, odd frames are dropped, generating a frame with
13597 unchanged height at half frame rate.
13602 Frame 1 Frame 2 Frame 3 Frame 4
13604 11111 22222 33333 44444
13605 11111 22222 33333 44444
13606 11111 22222 33333 44444
13607 11111 22222 33333 44444
13617 Expand each frame to full height, but pad alternate lines with black,
13618 generating a frame with double height at the same input frame rate.
13623 Frame 1 Frame 2 Frame 3 Frame 4
13625 11111 22222 33333 44444
13626 11111 22222 33333 44444
13627 11111 22222 33333 44444
13628 11111 22222 33333 44444
13631 11111 ..... 33333 .....
13632 ..... 22222 ..... 44444
13633 11111 ..... 33333 .....
13634 ..... 22222 ..... 44444
13635 11111 ..... 33333 .....
13636 ..... 22222 ..... 44444
13637 11111 ..... 33333 .....
13638 ..... 22222 ..... 44444
13642 @item interleave_top, 4
13643 Interleave the upper field from odd frames with the lower field from
13644 even frames, generating a frame with unchanged height at half frame rate.
13649 Frame 1 Frame 2 Frame 3 Frame 4
13651 11111<- 22222 33333<- 44444
13652 11111 22222<- 33333 44444<-
13653 11111<- 22222 33333<- 44444
13654 11111 22222<- 33333 44444<-
13664 @item interleave_bottom, 5
13665 Interleave the lower field from odd frames with the upper field from
13666 even frames, generating a frame with unchanged height at half frame rate.
13671 Frame 1 Frame 2 Frame 3 Frame 4
13673 11111 22222<- 33333 44444<-
13674 11111<- 22222 33333<- 44444
13675 11111 22222<- 33333 44444<-
13676 11111<- 22222 33333<- 44444
13686 @item interlacex2, 6
13687 Double frame rate with unchanged height. Frames are inserted each
13688 containing the second temporal field from the previous input frame and
13689 the first temporal field from the next input frame. This mode relies on
13690 the top_field_first flag. Useful for interlaced video displays with no
13691 field synchronisation.
13696 Frame 1 Frame 2 Frame 3 Frame 4
13698 11111 22222 33333 44444
13699 11111 22222 33333 44444
13700 11111 22222 33333 44444
13701 11111 22222 33333 44444
13704 11111 22222 22222 33333 33333 44444 44444
13705 11111 11111 22222 22222 33333 33333 44444
13706 11111 22222 22222 33333 33333 44444 44444
13707 11111 11111 22222 22222 33333 33333 44444
13712 Move odd frames into the upper field, even into the lower field,
13713 generating a double height frame at same frame rate.
13718 Frame 1 Frame 2 Frame 3 Frame 4
13720 11111 22222 33333 44444
13721 11111 22222 33333 44444
13722 11111 22222 33333 44444
13723 11111 22222 33333 44444
13726 11111 33333 33333 55555
13727 22222 22222 44444 44444
13728 11111 33333 33333 55555
13729 22222 22222 44444 44444
13730 11111 33333 33333 55555
13731 22222 22222 44444 44444
13732 11111 33333 33333 55555
13733 22222 22222 44444 44444
13738 Numeric values are deprecated but are accepted for backward
13739 compatibility reasons.
13741 Default mode is @code{merge}.
13744 Specify flags influencing the filter process.
13746 Available value for @var{flags} is:
13749 @item low_pass_filter, vlfp
13750 Enable vertical low-pass filtering in the filter.
13751 Vertical low-pass filtering is required when creating an interlaced
13752 destination from a progressive source which contains high-frequency
13753 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13756 Vertical low-pass filtering can only be enabled for @option{mode}
13757 @var{interleave_top} and @var{interleave_bottom}.
13764 Transpose rows with columns in the input video and optionally flip it.
13766 It accepts the following parameters:
13771 Specify the transposition direction.
13773 Can assume the following values:
13775 @item 0, 4, cclock_flip
13776 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13784 Rotate by 90 degrees clockwise, that is:
13792 Rotate by 90 degrees counterclockwise, that is:
13799 @item 3, 7, clock_flip
13800 Rotate by 90 degrees clockwise and vertically flip, that is:
13808 For values between 4-7, the transposition is only done if the input
13809 video geometry is portrait and not landscape. These values are
13810 deprecated, the @code{passthrough} option should be used instead.
13812 Numerical values are deprecated, and should be dropped in favor of
13813 symbolic constants.
13816 Do not apply the transposition if the input geometry matches the one
13817 specified by the specified value. It accepts the following values:
13820 Always apply transposition.
13822 Preserve portrait geometry (when @var{height} >= @var{width}).
13824 Preserve landscape geometry (when @var{width} >= @var{height}).
13827 Default value is @code{none}.
13830 For example to rotate by 90 degrees clockwise and preserve portrait
13833 transpose=dir=1:passthrough=portrait
13836 The command above can also be specified as:
13838 transpose=1:portrait
13842 Trim the input so that the output contains one continuous subpart of the input.
13844 It accepts the following parameters:
13847 Specify the time of the start of the kept section, i.e. the frame with the
13848 timestamp @var{start} will be the first frame in the output.
13851 Specify the time of the first frame that will be dropped, i.e. the frame
13852 immediately preceding the one with the timestamp @var{end} will be the last
13853 frame in the output.
13856 This is the same as @var{start}, except this option sets the start timestamp
13857 in timebase units instead of seconds.
13860 This is the same as @var{end}, except this option sets the end timestamp
13861 in timebase units instead of seconds.
13864 The maximum duration of the output in seconds.
13867 The number of the first frame that should be passed to the output.
13870 The number of the first frame that should be dropped.
13873 @option{start}, @option{end}, and @option{duration} are expressed as time
13874 duration specifications; see
13875 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13876 for the accepted syntax.
13878 Note that the first two sets of the start/end options and the @option{duration}
13879 option look at the frame timestamp, while the _frame variants simply count the
13880 frames that pass through the filter. Also note that this filter does not modify
13881 the timestamps. If you wish for the output timestamps to start at zero, insert a
13882 setpts filter after the trim filter.
13884 If multiple start or end options are set, this filter tries to be greedy and
13885 keep all the frames that match at least one of the specified constraints. To keep
13886 only the part that matches all the constraints at once, chain multiple trim
13889 The defaults are such that all the input is kept. So it is possible to set e.g.
13890 just the end values to keep everything before the specified time.
13895 Drop everything except the second minute of input:
13897 ffmpeg -i INPUT -vf trim=60:120
13901 Keep only the first second:
13903 ffmpeg -i INPUT -vf trim=duration=1
13912 Sharpen or blur the input video.
13914 It accepts the following parameters:
13917 @item luma_msize_x, lx
13918 Set the luma matrix horizontal size. It must be an odd integer between
13919 3 and 23. The default value is 5.
13921 @item luma_msize_y, ly
13922 Set the luma matrix vertical size. It must be an odd integer between 3
13923 and 23. The default value is 5.
13925 @item luma_amount, la
13926 Set the luma effect strength. It must be a floating point number, reasonable
13927 values lay between -1.5 and 1.5.
13929 Negative values will blur the input video, while positive values will
13930 sharpen it, a value of zero will disable the effect.
13932 Default value is 1.0.
13934 @item chroma_msize_x, cx
13935 Set the chroma matrix horizontal size. It must be an odd integer
13936 between 3 and 23. The default value is 5.
13938 @item chroma_msize_y, cy
13939 Set the chroma matrix vertical size. It must be an odd integer
13940 between 3 and 23. The default value is 5.
13942 @item chroma_amount, ca
13943 Set the chroma effect strength. It must be a floating point number, reasonable
13944 values lay between -1.5 and 1.5.
13946 Negative values will blur the input video, while positive values will
13947 sharpen it, a value of zero will disable the effect.
13949 Default value is 0.0.
13952 If set to 1, specify using OpenCL capabilities, only available if
13953 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13957 All parameters are optional and default to the equivalent of the
13958 string '5:5:1.0:5:5:0.0'.
13960 @subsection Examples
13964 Apply strong luma sharpen effect:
13966 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13970 Apply a strong blur of both luma and chroma parameters:
13972 unsharp=7:7:-2:7:7:-2
13978 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13979 the image at several (or - in the case of @option{quality} level @code{8} - all)
13980 shifts and average the results.
13982 The way this differs from the behavior of spp is that uspp actually encodes &
13983 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13984 DCT similar to MJPEG.
13986 The filter accepts the following options:
13990 Set quality. This option defines the number of levels for averaging. It accepts
13991 an integer in the range 0-8. If set to @code{0}, the filter will have no
13992 effect. A value of @code{8} means the higher quality. For each increment of
13993 that value the speed drops by a factor of approximately 2. Default value is
13997 Force a constant quantization parameter. If not set, the filter will use the QP
13998 from the video stream (if available).
14001 @section vaguedenoiser
14003 Apply a wavelet based denoiser.
14005 It transforms each frame from the video input into the wavelet domain,
14006 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
14007 the obtained coefficients. It does an inverse wavelet transform after.
14008 Due to wavelet properties, it should give a nice smoothed result, and
14009 reduced noise, without blurring picture features.
14011 This filter accepts the following options:
14015 The filtering strength. The higher, the more filtered the video will be.
14016 Hard thresholding can use a higher threshold than soft thresholding
14017 before the video looks overfiltered.
14020 The filtering method the filter will use.
14022 It accepts the following values:
14025 All values under the threshold will be zeroed.
14028 All values under the threshold will be zeroed. All values above will be
14029 reduced by the threshold.
14032 Scales or nullifies coefficients - intermediary between (more) soft and
14033 (less) hard thresholding.
14037 Number of times, the wavelet will decompose the picture. Picture can't
14038 be decomposed beyond a particular point (typically, 8 for a 640x480
14039 frame - as 2^9 = 512 > 480)
14042 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
14045 A list of the planes to process. By default all planes are processed.
14048 @section vectorscope
14050 Display 2 color component values in the two dimensional graph (which is called
14053 This filter accepts the following options:
14057 Set vectorscope mode.
14059 It accepts the following values:
14062 Gray values are displayed on graph, higher brightness means more pixels have
14063 same component color value on location in graph. This is the default mode.
14066 Gray values are displayed on graph. Surrounding pixels values which are not
14067 present in video frame are drawn in gradient of 2 color components which are
14068 set by option @code{x} and @code{y}. The 3rd color component is static.
14071 Actual color components values present in video frame are displayed on graph.
14074 Similar as color2 but higher frequency of same values @code{x} and @code{y}
14075 on graph increases value of another color component, which is luminance by
14076 default values of @code{x} and @code{y}.
14079 Actual colors present in video frame are displayed on graph. If two different
14080 colors map to same position on graph then color with higher value of component
14081 not present in graph is picked.
14084 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
14085 component picked from radial gradient.
14089 Set which color component will be represented on X-axis. Default is @code{1}.
14092 Set which color component will be represented on Y-axis. Default is @code{2}.
14095 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
14096 of color component which represents frequency of (X, Y) location in graph.
14101 No envelope, this is default.
14104 Instant envelope, even darkest single pixel will be clearly highlighted.
14107 Hold maximum and minimum values presented in graph over time. This way you
14108 can still spot out of range values without constantly looking at vectorscope.
14111 Peak and instant envelope combined together.
14115 Set what kind of graticule to draw.
14123 Set graticule opacity.
14126 Set graticule flags.
14130 Draw graticule for white point.
14133 Draw graticule for black point.
14136 Draw color points short names.
14140 Set background opacity.
14142 @item lthreshold, l
14143 Set low threshold for color component not represented on X or Y axis.
14144 Values lower than this value will be ignored. Default is 0.
14145 Note this value is multiplied with actual max possible value one pixel component
14146 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14149 @item hthreshold, h
14150 Set high threshold for color component not represented on X or Y axis.
14151 Values higher than this value will be ignored. Default is 1.
14152 Note this value is multiplied with actual max possible value one pixel component
14153 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14154 is 0.9 * 255 = 230.
14156 @item colorspace, c
14157 Set what kind of colorspace to use when drawing graticule.
14166 @anchor{vidstabdetect}
14167 @section vidstabdetect
14169 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14170 @ref{vidstabtransform} for pass 2.
14172 This filter generates a file with relative translation and rotation
14173 transform information about subsequent frames, which is then used by
14174 the @ref{vidstabtransform} filter.
14176 To enable compilation of this filter you need to configure FFmpeg with
14177 @code{--enable-libvidstab}.
14179 This filter accepts the following options:
14183 Set the path to the file used to write the transforms information.
14184 Default value is @file{transforms.trf}.
14187 Set how shaky the video is and how quick the camera is. It accepts an
14188 integer in the range 1-10, a value of 1 means little shakiness, a
14189 value of 10 means strong shakiness. Default value is 5.
14192 Set the accuracy of the detection process. It must be a value in the
14193 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14194 accuracy. Default value is 15.
14197 Set stepsize of the search process. The region around minimum is
14198 scanned with 1 pixel resolution. Default value is 6.
14201 Set minimum contrast. Below this value a local measurement field is
14202 discarded. Must be a floating point value in the range 0-1. Default
14206 Set reference frame number for tripod mode.
14208 If enabled, the motion of the frames is compared to a reference frame
14209 in the filtered stream, identified by the specified number. The idea
14210 is to compensate all movements in a more-or-less static scene and keep
14211 the camera view absolutely still.
14213 If set to 0, it is disabled. The frames are counted starting from 1.
14216 Show fields and transforms in the resulting frames. It accepts an
14217 integer in the range 0-2. Default value is 0, which disables any
14221 @subsection Examples
14225 Use default values:
14231 Analyze strongly shaky movie and put the results in file
14232 @file{mytransforms.trf}:
14234 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14238 Visualize the result of internal transformations in the resulting
14241 vidstabdetect=show=1
14245 Analyze a video with medium shakiness using @command{ffmpeg}:
14247 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14251 @anchor{vidstabtransform}
14252 @section vidstabtransform
14254 Video stabilization/deshaking: pass 2 of 2,
14255 see @ref{vidstabdetect} for pass 1.
14257 Read a file with transform information for each frame and
14258 apply/compensate them. Together with the @ref{vidstabdetect}
14259 filter this can be used to deshake videos. See also
14260 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14261 the @ref{unsharp} filter, see below.
14263 To enable compilation of this filter you need to configure FFmpeg with
14264 @code{--enable-libvidstab}.
14266 @subsection Options
14270 Set path to the file used to read the transforms. Default value is
14271 @file{transforms.trf}.
14274 Set the number of frames (value*2 + 1) used for lowpass filtering the
14275 camera movements. Default value is 10.
14277 For example a number of 10 means that 21 frames are used (10 in the
14278 past and 10 in the future) to smoothen the motion in the video. A
14279 larger value leads to a smoother video, but limits the acceleration of
14280 the camera (pan/tilt movements). 0 is a special case where a static
14281 camera is simulated.
14284 Set the camera path optimization algorithm.
14286 Accepted values are:
14289 gaussian kernel low-pass filter on camera motion (default)
14291 averaging on transformations
14295 Set maximal number of pixels to translate frames. Default value is -1,
14299 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14300 value is -1, meaning no limit.
14303 Specify how to deal with borders that may be visible due to movement
14306 Available values are:
14309 keep image information from previous frame (default)
14311 fill the border black
14315 Invert transforms if set to 1. Default value is 0.
14318 Consider transforms as relative to previous frame if set to 1,
14319 absolute if set to 0. Default value is 0.
14322 Set percentage to zoom. A positive value will result in a zoom-in
14323 effect, a negative value in a zoom-out effect. Default value is 0 (no
14327 Set optimal zooming to avoid borders.
14329 Accepted values are:
14334 optimal static zoom value is determined (only very strong movements
14335 will lead to visible borders) (default)
14337 optimal adaptive zoom value is determined (no borders will be
14338 visible), see @option{zoomspeed}
14341 Note that the value given at zoom is added to the one calculated here.
14344 Set percent to zoom maximally each frame (enabled when
14345 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14349 Specify type of interpolation.
14351 Available values are:
14356 linear only horizontal
14358 linear in both directions (default)
14360 cubic in both directions (slow)
14364 Enable virtual tripod mode if set to 1, which is equivalent to
14365 @code{relative=0:smoothing=0}. Default value is 0.
14367 Use also @code{tripod} option of @ref{vidstabdetect}.
14370 Increase log verbosity if set to 1. Also the detected global motions
14371 are written to the temporary file @file{global_motions.trf}. Default
14375 @subsection Examples
14379 Use @command{ffmpeg} for a typical stabilization with default values:
14381 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14384 Note the use of the @ref{unsharp} filter which is always recommended.
14387 Zoom in a bit more and load transform data from a given file:
14389 vidstabtransform=zoom=5:input="mytransforms.trf"
14393 Smoothen the video even more:
14395 vidstabtransform=smoothing=30
14401 Flip the input video vertically.
14403 For example, to vertically flip a video with @command{ffmpeg}:
14405 ffmpeg -i in.avi -vf "vflip" out.avi
14411 Make or reverse a natural vignetting effect.
14413 The filter accepts the following options:
14417 Set lens angle expression as a number of radians.
14419 The value is clipped in the @code{[0,PI/2]} range.
14421 Default value: @code{"PI/5"}
14425 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14429 Set forward/backward mode.
14431 Available modes are:
14434 The larger the distance from the central point, the darker the image becomes.
14437 The larger the distance from the central point, the brighter the image becomes.
14438 This can be used to reverse a vignette effect, though there is no automatic
14439 detection to extract the lens @option{angle} and other settings (yet). It can
14440 also be used to create a burning effect.
14443 Default value is @samp{forward}.
14446 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14448 It accepts the following values:
14451 Evaluate expressions only once during the filter initialization.
14454 Evaluate expressions for each incoming frame. This is way slower than the
14455 @samp{init} mode since it requires all the scalers to be re-computed, but it
14456 allows advanced dynamic expressions.
14459 Default value is @samp{init}.
14462 Set dithering to reduce the circular banding effects. Default is @code{1}
14466 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14467 Setting this value to the SAR of the input will make a rectangular vignetting
14468 following the dimensions of the video.
14470 Default is @code{1/1}.
14473 @subsection Expressions
14475 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14476 following parameters.
14481 input width and height
14484 the number of input frame, starting from 0
14487 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14488 @var{TB} units, NAN if undefined
14491 frame rate of the input video, NAN if the input frame rate is unknown
14494 the PTS (Presentation TimeStamp) of the filtered video frame,
14495 expressed in seconds, NAN if undefined
14498 time base of the input video
14502 @subsection Examples
14506 Apply simple strong vignetting effect:
14512 Make a flickering vignetting:
14514 vignette='PI/4+random(1)*PI/50':eval=frame
14520 Stack input videos vertically.
14522 All streams must be of same pixel format and of same width.
14524 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14525 to create same output.
14527 The filter accept the following option:
14531 Set number of input streams. Default is 2.
14534 If set to 1, force the output to terminate when the shortest input
14535 terminates. Default value is 0.
14540 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14541 Deinterlacing Filter").
14543 Based on the process described by Martin Weston for BBC R&D, and
14544 implemented based on the de-interlace algorithm written by Jim
14545 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14546 uses filter coefficients calculated by BBC R&D.
14548 There are two sets of filter coefficients, so called "simple":
14549 and "complex". Which set of filter coefficients is used can
14550 be set by passing an optional parameter:
14554 Set the interlacing filter coefficients. Accepts one of the following values:
14558 Simple filter coefficient set.
14560 More-complex filter coefficient set.
14562 Default value is @samp{complex}.
14565 Specify which frames to deinterlace. Accept one of the following values:
14569 Deinterlace all frames,
14571 Only deinterlace frames marked as interlaced.
14574 Default value is @samp{all}.
14578 Video waveform monitor.
14580 The waveform monitor plots color component intensity. By default luminance
14581 only. Each column of the waveform corresponds to a column of pixels in the
14584 It accepts the following options:
14588 Can be either @code{row}, or @code{column}. Default is @code{column}.
14589 In row mode, the graph on the left side represents color component value 0 and
14590 the right side represents value = 255. In column mode, the top side represents
14591 color component value = 0 and bottom side represents value = 255.
14594 Set intensity. Smaller values are useful to find out how many values of the same
14595 luminance are distributed across input rows/columns.
14596 Default value is @code{0.04}. Allowed range is [0, 1].
14599 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14600 In mirrored mode, higher values will be represented on the left
14601 side for @code{row} mode and at the top for @code{column} mode. Default is
14602 @code{1} (mirrored).
14606 It accepts the following values:
14609 Presents information identical to that in the @code{parade}, except
14610 that the graphs representing color components are superimposed directly
14613 This display mode makes it easier to spot relative differences or similarities
14614 in overlapping areas of the color components that are supposed to be identical,
14615 such as neutral whites, grays, or blacks.
14618 Display separate graph for the color components side by side in
14619 @code{row} mode or one below the other in @code{column} mode.
14622 Display separate graph for the color components side by side in
14623 @code{column} mode or one below the other in @code{row} mode.
14625 Using this display mode makes it easy to spot color casts in the highlights
14626 and shadows of an image, by comparing the contours of the top and the bottom
14627 graphs of each waveform. Since whites, grays, and blacks are characterized
14628 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14629 should display three waveforms of roughly equal width/height. If not, the
14630 correction is easy to perform by making level adjustments the three waveforms.
14632 Default is @code{stack}.
14634 @item components, c
14635 Set which color components to display. Default is 1, which means only luminance
14636 or red color component if input is in RGB colorspace. If is set for example to
14637 7 it will display all 3 (if) available color components.
14642 No envelope, this is default.
14645 Instant envelope, minimum and maximum values presented in graph will be easily
14646 visible even with small @code{step} value.
14649 Hold minimum and maximum values presented in graph across time. This way you
14650 can still spot out of range values without constantly looking at waveforms.
14653 Peak and instant envelope combined together.
14659 No filtering, this is default.
14662 Luma and chroma combined together.
14665 Similar as above, but shows difference between blue and red chroma.
14668 Displays only chroma.
14671 Displays actual color value on waveform.
14674 Similar as above, but with luma showing frequency of chroma values.
14678 Set which graticule to display.
14682 Do not display graticule.
14685 Display green graticule showing legal broadcast ranges.
14689 Set graticule opacity.
14692 Set graticule flags.
14696 Draw numbers above lines. By default enabled.
14699 Draw dots instead of lines.
14703 Set scale used for displaying graticule.
14710 Default is digital.
14713 Set background opacity.
14716 @section weave, doubleweave
14718 The @code{weave} takes a field-based video input and join
14719 each two sequential fields into single frame, producing a new double
14720 height clip with half the frame rate and half the frame count.
14722 The @code{doubleweave} works same as @code{weave} but without
14723 halving frame rate and frame count.
14725 It accepts the following option:
14729 Set first field. Available values are:
14733 Set the frame as top-field-first.
14736 Set the frame as bottom-field-first.
14740 @subsection Examples
14744 Interlace video using @ref{select} and @ref{separatefields} filter:
14746 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14751 Apply the xBR high-quality magnification filter which is designed for pixel
14752 art. It follows a set of edge-detection rules, see
14753 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14755 It accepts the following option:
14759 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14760 @code{3xBR} and @code{4} for @code{4xBR}.
14761 Default is @code{3}.
14767 Deinterlace the input video ("yadif" means "yet another deinterlacing
14770 It accepts the following parameters:
14776 The interlacing mode to adopt. It accepts one of the following values:
14779 @item 0, send_frame
14780 Output one frame for each frame.
14781 @item 1, send_field
14782 Output one frame for each field.
14783 @item 2, send_frame_nospatial
14784 Like @code{send_frame}, but it skips the spatial interlacing check.
14785 @item 3, send_field_nospatial
14786 Like @code{send_field}, but it skips the spatial interlacing check.
14789 The default value is @code{send_frame}.
14792 The picture field parity assumed for the input interlaced video. It accepts one
14793 of the following values:
14797 Assume the top field is first.
14799 Assume the bottom field is first.
14801 Enable automatic detection of field parity.
14804 The default value is @code{auto}.
14805 If the interlacing is unknown or the decoder does not export this information,
14806 top field first will be assumed.
14809 Specify which frames to deinterlace. Accept one of the following
14814 Deinterlace all frames.
14815 @item 1, interlaced
14816 Only deinterlace frames marked as interlaced.
14819 The default value is @code{all}.
14824 Apply Zoom & Pan effect.
14826 This filter accepts the following options:
14830 Set the zoom expression. Default is 1.
14834 Set the x and y expression. Default is 0.
14837 Set the duration expression in number of frames.
14838 This sets for how many number of frames effect will last for
14839 single input image.
14842 Set the output image size, default is 'hd720'.
14845 Set the output frame rate, default is '25'.
14848 Each expression can contain the following constants:
14867 Output frame count.
14871 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14872 for current input frame.
14876 'x' and 'y' of last output frame of previous input frame or 0 when there was
14877 not yet such frame (first input frame).
14880 Last calculated zoom from 'z' expression for current input frame.
14883 Last calculated zoom of last output frame of previous input frame.
14886 Number of output frames for current input frame. Calculated from 'd' expression
14887 for each input frame.
14890 number of output frames created for previous input frame
14893 Rational number: input width / input height
14896 sample aspect ratio
14899 display aspect ratio
14903 @subsection Examples
14907 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14909 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
14913 Zoom-in up to 1.5 and pan always at center of picture:
14915 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14919 Same as above but without pausing:
14921 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14926 Scale (resize) the input video, using the z.lib library:
14927 https://github.com/sekrit-twc/zimg.
14929 The zscale filter forces the output display aspect ratio to be the same
14930 as the input, by changing the output sample aspect ratio.
14932 If the input image format is different from the format requested by
14933 the next filter, the zscale filter will convert the input to the
14936 @subsection Options
14937 The filter accepts the following options.
14942 Set the output video dimension expression. Default value is the input
14945 If the @var{width} or @var{w} is 0, the input width is used for the output.
14946 If the @var{height} or @var{h} is 0, the input height is used for the output.
14948 If one of the values is -1, the zscale filter will use a value that
14949 maintains the aspect ratio of the input image, calculated from the
14950 other specified dimension. If both of them are -1, the input size is
14953 If one of the values is -n with n > 1, the zscale filter will also use a value
14954 that maintains the aspect ratio of the input image, calculated from the other
14955 specified dimension. After that it will, however, make sure that the calculated
14956 dimension is divisible by n and adjust the value if necessary.
14958 See below for the list of accepted constants for use in the dimension
14962 Set the video size. For the syntax of this option, check the
14963 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14966 Set the dither type.
14968 Possible values are:
14973 @item error_diffusion
14979 Set the resize filter type.
14981 Possible values are:
14991 Default is bilinear.
14994 Set the color range.
14996 Possible values are:
15003 Default is same as input.
15006 Set the color primaries.
15008 Possible values are:
15018 Default is same as input.
15021 Set the transfer characteristics.
15023 Possible values are:
15037 Default is same as input.
15040 Set the colorspace matrix.
15042 Possible value are:
15053 Default is same as input.
15056 Set the input color range.
15058 Possible values are:
15065 Default is same as input.
15067 @item primariesin, pin
15068 Set the input color primaries.
15070 Possible values are:
15080 Default is same as input.
15082 @item transferin, tin
15083 Set the input transfer characteristics.
15085 Possible values are:
15096 Default is same as input.
15098 @item matrixin, min
15099 Set the input colorspace matrix.
15101 Possible value are:
15113 Set the output chroma location.
15115 Possible values are:
15126 @item chromalin, cin
15127 Set the input chroma location.
15129 Possible values are:
15141 Set the nominal peak luminance.
15144 The values of the @option{w} and @option{h} options are expressions
15145 containing the following constants:
15150 The input width and height
15154 These are the same as @var{in_w} and @var{in_h}.
15158 The output (scaled) width and height
15162 These are the same as @var{out_w} and @var{out_h}
15165 The same as @var{iw} / @var{ih}
15168 input sample aspect ratio
15171 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15175 horizontal and vertical input chroma subsample values. For example for the
15176 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15180 horizontal and vertical output chroma subsample values. For example for the
15181 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15187 @c man end VIDEO FILTERS
15189 @chapter Video Sources
15190 @c man begin VIDEO SOURCES
15192 Below is a description of the currently available video sources.
15196 Buffer video frames, and make them available to the filter chain.
15198 This source is mainly intended for a programmatic use, in particular
15199 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15201 It accepts the following parameters:
15206 Specify the size (width and height) of the buffered video frames. For the
15207 syntax of this option, check the
15208 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15211 The input video width.
15214 The input video height.
15217 A string representing the pixel format of the buffered video frames.
15218 It may be a number corresponding to a pixel format, or a pixel format
15222 Specify the timebase assumed by the timestamps of the buffered frames.
15225 Specify the frame rate expected for the video stream.
15227 @item pixel_aspect, sar
15228 The sample (pixel) aspect ratio of the input video.
15231 Specify the optional parameters to be used for the scale filter which
15232 is automatically inserted when an input change is detected in the
15233 input size or format.
15235 @item hw_frames_ctx
15236 When using a hardware pixel format, this should be a reference to an
15237 AVHWFramesContext describing input frames.
15242 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15245 will instruct the source to accept video frames with size 320x240 and
15246 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15247 square pixels (1:1 sample aspect ratio).
15248 Since the pixel format with name "yuv410p" corresponds to the number 6
15249 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15250 this example corresponds to:
15252 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15255 Alternatively, the options can be specified as a flat string, but this
15256 syntax is deprecated:
15258 @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}]
15262 Create a pattern generated by an elementary cellular automaton.
15264 The initial state of the cellular automaton can be defined through the
15265 @option{filename} and @option{pattern} options. If such options are
15266 not specified an initial state is created randomly.
15268 At each new frame a new row in the video is filled with the result of
15269 the cellular automaton next generation. The behavior when the whole
15270 frame is filled is defined by the @option{scroll} option.
15272 This source accepts the following options:
15276 Read the initial cellular automaton state, i.e. the starting row, from
15277 the specified file.
15278 In the file, each non-whitespace character is considered an alive
15279 cell, a newline will terminate the row, and further characters in the
15280 file will be ignored.
15283 Read the initial cellular automaton state, i.e. the starting row, from
15284 the specified string.
15286 Each non-whitespace character in the string is considered an alive
15287 cell, a newline will terminate the row, and further characters in the
15288 string will be ignored.
15291 Set the video rate, that is the number of frames generated per second.
15294 @item random_fill_ratio, ratio
15295 Set the random fill ratio for the initial cellular automaton row. It
15296 is a floating point number value ranging from 0 to 1, defaults to
15299 This option is ignored when a file or a pattern is specified.
15301 @item random_seed, seed
15302 Set the seed for filling randomly the initial row, must be an integer
15303 included between 0 and UINT32_MAX. If not specified, or if explicitly
15304 set to -1, the filter will try to use a good random seed on a best
15308 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15309 Default value is 110.
15312 Set the size of the output video. For the syntax of this option, check the
15313 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15315 If @option{filename} or @option{pattern} is specified, the size is set
15316 by default to the width of the specified initial state row, and the
15317 height is set to @var{width} * PHI.
15319 If @option{size} is set, it must contain the width of the specified
15320 pattern string, and the specified pattern will be centered in the
15323 If a filename or a pattern string is not specified, the size value
15324 defaults to "320x518" (used for a randomly generated initial state).
15327 If set to 1, scroll the output upward when all the rows in the output
15328 have been already filled. If set to 0, the new generated row will be
15329 written over the top row just after the bottom row is filled.
15332 @item start_full, full
15333 If set to 1, completely fill the output with generated rows before
15334 outputting the first frame.
15335 This is the default behavior, for disabling set the value to 0.
15338 If set to 1, stitch the left and right row edges together.
15339 This is the default behavior, for disabling set the value to 0.
15342 @subsection Examples
15346 Read the initial state from @file{pattern}, and specify an output of
15349 cellauto=f=pattern:s=200x400
15353 Generate a random initial row with a width of 200 cells, with a fill
15356 cellauto=ratio=2/3:s=200x200
15360 Create a pattern generated by rule 18 starting by a single alive cell
15361 centered on an initial row with width 100:
15363 cellauto=p=@@:s=100x400:full=0:rule=18
15367 Specify a more elaborated initial pattern:
15369 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15374 @anchor{coreimagesrc}
15375 @section coreimagesrc
15376 Video source generated on GPU using Apple's CoreImage API on OSX.
15378 This video source is a specialized version of the @ref{coreimage} video filter.
15379 Use a core image generator at the beginning of the applied filterchain to
15380 generate the content.
15382 The coreimagesrc video source accepts the following options:
15384 @item list_generators
15385 List all available generators along with all their respective options as well as
15386 possible minimum and maximum values along with the default values.
15388 list_generators=true
15392 Specify the size of the sourced video. For the syntax of this option, check the
15393 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15394 The default value is @code{320x240}.
15397 Specify the frame rate of the sourced video, as the number of frames
15398 generated per second. It has to be a string in the format
15399 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15400 number or a valid video frame rate abbreviation. The default value is
15404 Set the sample aspect ratio of the sourced video.
15407 Set the duration of the sourced video. See
15408 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15409 for the accepted syntax.
15411 If not specified, or the expressed duration is negative, the video is
15412 supposed to be generated forever.
15415 Additionally, all options of the @ref{coreimage} video filter are accepted.
15416 A complete filterchain can be used for further processing of the
15417 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15418 and examples for details.
15420 @subsection Examples
15425 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15426 given as complete and escaped command-line for Apple's standard bash shell:
15428 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15430 This example is equivalent to the QRCode example of @ref{coreimage} without the
15431 need for a nullsrc video source.
15435 @section mandelbrot
15437 Generate a Mandelbrot set fractal, and progressively zoom towards the
15438 point specified with @var{start_x} and @var{start_y}.
15440 This source accepts the following options:
15445 Set the terminal pts value. Default value is 400.
15448 Set the terminal scale value.
15449 Must be a floating point value. Default value is 0.3.
15452 Set the inner coloring mode, that is the algorithm used to draw the
15453 Mandelbrot fractal internal region.
15455 It shall assume one of the following values:
15460 Show time until convergence.
15462 Set color based on point closest to the origin of the iterations.
15467 Default value is @var{mincol}.
15470 Set the bailout value. Default value is 10.0.
15473 Set the maximum of iterations performed by the rendering
15474 algorithm. Default value is 7189.
15477 Set outer coloring mode.
15478 It shall assume one of following values:
15480 @item iteration_count
15481 Set iteration cound mode.
15482 @item normalized_iteration_count
15483 set normalized iteration count mode.
15485 Default value is @var{normalized_iteration_count}.
15488 Set frame rate, expressed as number of frames per second. Default
15492 Set frame size. For the syntax of this option, check the "Video
15493 size" section in the ffmpeg-utils manual. Default value is "640x480".
15496 Set the initial scale value. Default value is 3.0.
15499 Set the initial x position. Must be a floating point value between
15500 -100 and 100. Default value is -0.743643887037158704752191506114774.
15503 Set the initial y position. Must be a floating point value between
15504 -100 and 100. Default value is -0.131825904205311970493132056385139.
15509 Generate various test patterns, as generated by the MPlayer test filter.
15511 The size of the generated video is fixed, and is 256x256.
15512 This source is useful in particular for testing encoding features.
15514 This source accepts the following options:
15519 Specify the frame rate of the sourced video, as the number of frames
15520 generated per second. It has to be a string in the format
15521 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15522 number or a valid video frame rate abbreviation. The default value is
15526 Set the duration of the sourced video. See
15527 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15528 for the accepted syntax.
15530 If not specified, or the expressed duration is negative, the video is
15531 supposed to be generated forever.
15535 Set the number or the name of the test to perform. Supported tests are:
15551 Default value is "all", which will cycle through the list of all tests.
15556 mptestsrc=t=dc_luma
15559 will generate a "dc_luma" test pattern.
15561 @section frei0r_src
15563 Provide a frei0r source.
15565 To enable compilation of this filter you need to install the frei0r
15566 header and configure FFmpeg with @code{--enable-frei0r}.
15568 This source accepts the following parameters:
15573 The size of the video to generate. For the syntax of this option, check the
15574 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15577 The framerate of the generated video. It may be a string of the form
15578 @var{num}/@var{den} or a frame rate abbreviation.
15581 The name to the frei0r source to load. For more information regarding frei0r and
15582 how to set the parameters, read the @ref{frei0r} section in the video filters
15585 @item filter_params
15586 A '|'-separated list of parameters to pass to the frei0r source.
15590 For example, to generate a frei0r partik0l source with size 200x200
15591 and frame rate 10 which is overlaid on the overlay filter main input:
15593 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15598 Generate a life pattern.
15600 This source is based on a generalization of John Conway's life game.
15602 The sourced input represents a life grid, each pixel represents a cell
15603 which can be in one of two possible states, alive or dead. Every cell
15604 interacts with its eight neighbours, which are the cells that are
15605 horizontally, vertically, or diagonally adjacent.
15607 At each interaction the grid evolves according to the adopted rule,
15608 which specifies the number of neighbor alive cells which will make a
15609 cell stay alive or born. The @option{rule} option allows one to specify
15612 This source accepts the following options:
15616 Set the file from which to read the initial grid state. In the file,
15617 each non-whitespace character is considered an alive cell, and newline
15618 is used to delimit the end of each row.
15620 If this option is not specified, the initial grid is generated
15624 Set the video rate, that is the number of frames generated per second.
15627 @item random_fill_ratio, ratio
15628 Set the random fill ratio for the initial random grid. It is a
15629 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15630 It is ignored when a file is specified.
15632 @item random_seed, seed
15633 Set the seed for filling the initial random grid, must be an integer
15634 included between 0 and UINT32_MAX. If not specified, or if explicitly
15635 set to -1, the filter will try to use a good random seed on a best
15641 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15642 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15643 @var{NS} specifies the number of alive neighbor cells which make a
15644 live cell stay alive, and @var{NB} the number of alive neighbor cells
15645 which make a dead cell to become alive (i.e. to "born").
15646 "s" and "b" can be used in place of "S" and "B", respectively.
15648 Alternatively a rule can be specified by an 18-bits integer. The 9
15649 high order bits are used to encode the next cell state if it is alive
15650 for each number of neighbor alive cells, the low order bits specify
15651 the rule for "borning" new cells. Higher order bits encode for an
15652 higher number of neighbor cells.
15653 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15654 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15656 Default value is "S23/B3", which is the original Conway's game of life
15657 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15658 cells, and will born a new cell if there are three alive cells around
15662 Set the size of the output video. For the syntax of this option, check the
15663 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15665 If @option{filename} is specified, the size is set by default to the
15666 same size of the input file. If @option{size} is set, it must contain
15667 the size specified in the input file, and the initial grid defined in
15668 that file is centered in the larger resulting area.
15670 If a filename is not specified, the size value defaults to "320x240"
15671 (used for a randomly generated initial grid).
15674 If set to 1, stitch the left and right grid edges together, and the
15675 top and bottom edges also. Defaults to 1.
15678 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15679 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15680 value from 0 to 255.
15683 Set the color of living (or new born) cells.
15686 Set the color of dead cells. If @option{mold} is set, this is the first color
15687 used to represent a dead cell.
15690 Set mold color, for definitely dead and moldy cells.
15692 For the syntax of these 3 color options, check the "Color" section in the
15693 ffmpeg-utils manual.
15696 @subsection Examples
15700 Read a grid from @file{pattern}, and center it on a grid of size
15703 life=f=pattern:s=300x300
15707 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15709 life=ratio=2/3:s=200x200
15713 Specify a custom rule for evolving a randomly generated grid:
15719 Full example with slow death effect (mold) using @command{ffplay}:
15721 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15728 @anchor{haldclutsrc}
15730 @anchor{rgbtestsrc}
15732 @anchor{smptehdbars}
15735 @anchor{yuvtestsrc}
15736 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15738 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15740 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15742 The @code{color} source provides an uniformly colored input.
15744 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15745 @ref{haldclut} filter.
15747 The @code{nullsrc} source returns unprocessed video frames. It is
15748 mainly useful to be employed in analysis / debugging tools, or as the
15749 source for filters which ignore the input data.
15751 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15752 detecting RGB vs BGR issues. You should see a red, green and blue
15753 stripe from top to bottom.
15755 The @code{smptebars} source generates a color bars pattern, based on
15756 the SMPTE Engineering Guideline EG 1-1990.
15758 The @code{smptehdbars} source generates a color bars pattern, based on
15759 the SMPTE RP 219-2002.
15761 The @code{testsrc} source generates a test video pattern, showing a
15762 color pattern, a scrolling gradient and a timestamp. This is mainly
15763 intended for testing purposes.
15765 The @code{testsrc2} source is similar to testsrc, but supports more
15766 pixel formats instead of just @code{rgb24}. This allows using it as an
15767 input for other tests without requiring a format conversion.
15769 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15770 see a y, cb and cr stripe from top to bottom.
15772 The sources accept the following parameters:
15777 Specify the color of the source, only available in the @code{color}
15778 source. For the syntax of this option, check the "Color" section in the
15779 ffmpeg-utils manual.
15782 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15783 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15784 pixels to be used as identity matrix for 3D lookup tables. Each component is
15785 coded on a @code{1/(N*N)} scale.
15788 Specify the size of the sourced video. For the syntax of this option, check the
15789 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15790 The default value is @code{320x240}.
15792 This option is not available with the @code{haldclutsrc} filter.
15795 Specify the frame rate of the sourced video, as the number of frames
15796 generated per second. It has to be a string in the format
15797 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15798 number or a valid video frame rate abbreviation. The default value is
15802 Set the sample aspect ratio of the sourced video.
15805 Set the duration of the sourced video. See
15806 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15807 for the accepted syntax.
15809 If not specified, or the expressed duration is negative, the video is
15810 supposed to be generated forever.
15813 Set the number of decimals to show in the timestamp, only available in the
15814 @code{testsrc} source.
15816 The displayed timestamp value will correspond to the original
15817 timestamp value multiplied by the power of 10 of the specified
15818 value. Default value is 0.
15821 For example the following:
15823 testsrc=duration=5.3:size=qcif:rate=10
15826 will generate a video with a duration of 5.3 seconds, with size
15827 176x144 and a frame rate of 10 frames per second.
15829 The following graph description will generate a red source
15830 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15833 color=c=red@@0.2:s=qcif:r=10
15836 If the input content is to be ignored, @code{nullsrc} can be used. The
15837 following command generates noise in the luminance plane by employing
15838 the @code{geq} filter:
15840 nullsrc=s=256x256, geq=random(1)*255:128:128
15843 @subsection Commands
15845 The @code{color} source supports the following commands:
15849 Set the color of the created image. Accepts the same syntax of the
15850 corresponding @option{color} option.
15853 @c man end VIDEO SOURCES
15855 @chapter Video Sinks
15856 @c man begin VIDEO SINKS
15858 Below is a description of the currently available video sinks.
15860 @section buffersink
15862 Buffer video frames, and make them available to the end of the filter
15865 This sink is mainly intended for programmatic use, in particular
15866 through the interface defined in @file{libavfilter/buffersink.h}
15867 or the options system.
15869 It accepts a pointer to an AVBufferSinkContext structure, which
15870 defines the incoming buffers' formats, to be passed as the opaque
15871 parameter to @code{avfilter_init_filter} for initialization.
15875 Null video sink: do absolutely nothing with the input video. It is
15876 mainly useful as a template and for use in analysis / debugging
15879 @c man end VIDEO SINKS
15881 @chapter Multimedia Filters
15882 @c man begin MULTIMEDIA FILTERS
15884 Below is a description of the currently available multimedia filters.
15888 Convert input audio to a video output, displaying the audio bit scope.
15890 The filter accepts the following options:
15894 Set frame rate, expressed as number of frames per second. Default
15898 Specify the video size for the output. For the syntax of this option, check the
15899 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15900 Default value is @code{1024x256}.
15903 Specify list of colors separated by space or by '|' which will be used to
15904 draw channels. Unrecognized or missing colors will be replaced
15908 @section ahistogram
15910 Convert input audio to a video output, displaying the volume histogram.
15912 The filter accepts the following options:
15916 Specify how histogram is calculated.
15918 It accepts the following values:
15921 Use single histogram for all channels.
15923 Use separate histogram for each channel.
15925 Default is @code{single}.
15928 Set frame rate, expressed as number of frames per second. Default
15932 Specify the video size for the output. For the syntax of this option, check the
15933 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15934 Default value is @code{hd720}.
15939 It accepts the following values:
15950 reverse logarithmic
15952 Default is @code{log}.
15955 Set amplitude scale.
15957 It accepts the following values:
15964 Default is @code{log}.
15967 Set how much frames to accumulate in histogram.
15968 Defauls is 1. Setting this to -1 accumulates all frames.
15971 Set histogram ratio of window height.
15974 Set sonogram sliding.
15976 It accepts the following values:
15979 replace old rows with new ones.
15981 scroll from top to bottom.
15983 Default is @code{replace}.
15986 @section aphasemeter
15988 Convert input audio to a video output, displaying the audio phase.
15990 The filter accepts the following options:
15994 Set the output frame rate. Default value is @code{25}.
15997 Set the video size for the output. For the syntax of this option, check the
15998 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15999 Default value is @code{800x400}.
16004 Specify the red, green, blue contrast. Default values are @code{2},
16005 @code{7} and @code{1}.
16006 Allowed range is @code{[0, 255]}.
16009 Set color which will be used for drawing median phase. If color is
16010 @code{none} which is default, no median phase value will be drawn.
16013 Enable video output. Default is enabled.
16016 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
16017 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
16018 The @code{-1} means left and right channels are completely out of phase and
16019 @code{1} means channels are in phase.
16021 @section avectorscope
16023 Convert input audio to a video output, representing the audio vector
16026 The filter is used to measure the difference between channels of stereo
16027 audio stream. A monoaural signal, consisting of identical left and right
16028 signal, results in straight vertical line. Any stereo separation is visible
16029 as a deviation from this line, creating a Lissajous figure.
16030 If the straight (or deviation from it) but horizontal line appears this
16031 indicates that the left and right channels are out of phase.
16033 The filter accepts the following options:
16037 Set the vectorscope mode.
16039 Available values are:
16042 Lissajous rotated by 45 degrees.
16045 Same as above but not rotated.
16048 Shape resembling half of circle.
16051 Default value is @samp{lissajous}.
16054 Set the video size for the output. For the syntax of this option, check the
16055 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16056 Default value is @code{400x400}.
16059 Set the output frame rate. Default value is @code{25}.
16065 Specify the red, green, blue and alpha contrast. Default values are @code{40},
16066 @code{160}, @code{80} and @code{255}.
16067 Allowed range is @code{[0, 255]}.
16073 Specify the red, green, blue and alpha fade. Default values are @code{15},
16074 @code{10}, @code{5} and @code{5}.
16075 Allowed range is @code{[0, 255]}.
16078 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
16081 Set the vectorscope drawing mode.
16083 Available values are:
16086 Draw dot for each sample.
16089 Draw line between previous and current sample.
16092 Default value is @samp{dot}.
16095 Specify amplitude scale of audio samples.
16097 Available values are:
16114 @subsection Examples
16118 Complete example using @command{ffplay}:
16120 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16121 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16125 @section bench, abench
16127 Benchmark part of a filtergraph.
16129 The filter accepts the following options:
16133 Start or stop a timer.
16135 Available values are:
16138 Get the current time, set it as frame metadata (using the key
16139 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
16142 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16143 the input frame metadata to get the time difference. Time difference, average,
16144 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16145 @code{min}) are then printed. The timestamps are expressed in seconds.
16149 @subsection Examples
16153 Benchmark @ref{selectivecolor} filter:
16155 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16161 Concatenate audio and video streams, joining them together one after the
16164 The filter works on segments of synchronized video and audio streams. All
16165 segments must have the same number of streams of each type, and that will
16166 also be the number of streams at output.
16168 The filter accepts the following options:
16173 Set the number of segments. Default is 2.
16176 Set the number of output video streams, that is also the number of video
16177 streams in each segment. Default is 1.
16180 Set the number of output audio streams, that is also the number of audio
16181 streams in each segment. Default is 0.
16184 Activate unsafe mode: do not fail if segments have a different format.
16188 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16189 @var{a} audio outputs.
16191 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16192 segment, in the same order as the outputs, then the inputs for the second
16195 Related streams do not always have exactly the same duration, for various
16196 reasons including codec frame size or sloppy authoring. For that reason,
16197 related synchronized streams (e.g. a video and its audio track) should be
16198 concatenated at once. The concat filter will use the duration of the longest
16199 stream in each segment (except the last one), and if necessary pad shorter
16200 audio streams with silence.
16202 For this filter to work correctly, all segments must start at timestamp 0.
16204 All corresponding streams must have the same parameters in all segments; the
16205 filtering system will automatically select a common pixel format for video
16206 streams, and a common sample format, sample rate and channel layout for
16207 audio streams, but other settings, such as resolution, must be converted
16208 explicitly by the user.
16210 Different frame rates are acceptable but will result in variable frame rate
16211 at output; be sure to configure the output file to handle it.
16213 @subsection Examples
16217 Concatenate an opening, an episode and an ending, all in bilingual version
16218 (video in stream 0, audio in streams 1 and 2):
16220 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16221 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16222 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16223 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16227 Concatenate two parts, handling audio and video separately, using the
16228 (a)movie sources, and adjusting the resolution:
16230 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16231 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16232 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16234 Note that a desync will happen at the stitch if the audio and video streams
16235 do not have exactly the same duration in the first file.
16239 @section drawgraph, adrawgraph
16241 Draw a graph using input video or audio metadata.
16243 It accepts the following parameters:
16247 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16250 Set 1st foreground color expression.
16253 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16256 Set 2nd foreground color expression.
16259 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16262 Set 3rd foreground color expression.
16265 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16268 Set 4th foreground color expression.
16271 Set minimal value of metadata value.
16274 Set maximal value of metadata value.
16277 Set graph background color. Default is white.
16282 Available values for mode is:
16289 Default is @code{line}.
16294 Available values for slide is:
16297 Draw new frame when right border is reached.
16300 Replace old columns with new ones.
16303 Scroll from right to left.
16306 Scroll from left to right.
16309 Draw single picture.
16312 Default is @code{frame}.
16315 Set size of graph video. For the syntax of this option, check the
16316 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16317 The default value is @code{900x256}.
16319 The foreground color expressions can use the following variables:
16322 Minimal value of metadata value.
16325 Maximal value of metadata value.
16328 Current metadata key value.
16331 The color is defined as 0xAABBGGRR.
16334 Example using metadata from @ref{signalstats} filter:
16336 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16339 Example using metadata from @ref{ebur128} filter:
16341 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16347 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16348 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16349 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16350 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16352 The filter also has a video output (see the @var{video} option) with a real
16353 time graph to observe the loudness evolution. The graphic contains the logged
16354 message mentioned above, so it is not printed anymore when this option is set,
16355 unless the verbose logging is set. The main graphing area contains the
16356 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16357 the momentary loudness (400 milliseconds).
16359 More information about the Loudness Recommendation EBU R128 on
16360 @url{http://tech.ebu.ch/loudness}.
16362 The filter accepts the following options:
16367 Activate the video output. The audio stream is passed unchanged whether this
16368 option is set or no. The video stream will be the first output stream if
16369 activated. Default is @code{0}.
16372 Set the video size. This option is for video only. For the syntax of this
16374 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16375 Default and minimum resolution is @code{640x480}.
16378 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16379 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16380 other integer value between this range is allowed.
16383 Set metadata injection. If set to @code{1}, the audio input will be segmented
16384 into 100ms output frames, each of them containing various loudness information
16385 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16387 Default is @code{0}.
16390 Force the frame logging level.
16392 Available values are:
16395 information logging level
16397 verbose logging level
16400 By default, the logging level is set to @var{info}. If the @option{video} or
16401 the @option{metadata} options are set, it switches to @var{verbose}.
16406 Available modes can be cumulated (the option is a @code{flag} type). Possible
16410 Disable any peak mode (default).
16412 Enable sample-peak mode.
16414 Simple peak mode looking for the higher sample value. It logs a message
16415 for sample-peak (identified by @code{SPK}).
16417 Enable true-peak mode.
16419 If enabled, the peak lookup is done on an over-sampled version of the input
16420 stream for better peak accuracy. It logs a message for true-peak.
16421 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16422 This mode requires a build with @code{libswresample}.
16426 Treat mono input files as "dual mono". If a mono file is intended for playback
16427 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16428 If set to @code{true}, this option will compensate for this effect.
16429 Multi-channel input files are not affected by this option.
16432 Set a specific pan law to be used for the measurement of dual mono files.
16433 This parameter is optional, and has a default value of -3.01dB.
16436 @subsection Examples
16440 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16442 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16446 Run an analysis with @command{ffmpeg}:
16448 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16452 @section interleave, ainterleave
16454 Temporally interleave frames from several inputs.
16456 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16458 These filters read frames from several inputs and send the oldest
16459 queued frame to the output.
16461 Input streams must have well defined, monotonically increasing frame
16464 In order to submit one frame to output, these filters need to enqueue
16465 at least one frame for each input, so they cannot work in case one
16466 input is not yet terminated and will not receive incoming frames.
16468 For example consider the case when one input is a @code{select} filter
16469 which always drops input frames. The @code{interleave} filter will keep
16470 reading from that input, but it will never be able to send new frames
16471 to output until the input sends an end-of-stream signal.
16473 Also, depending on inputs synchronization, the filters will drop
16474 frames in case one input receives more frames than the other ones, and
16475 the queue is already filled.
16477 These filters accept the following options:
16481 Set the number of different inputs, it is 2 by default.
16484 @subsection Examples
16488 Interleave frames belonging to different streams using @command{ffmpeg}:
16490 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16494 Add flickering blur effect:
16496 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16500 @section metadata, ametadata
16502 Manipulate frame metadata.
16504 This filter accepts the following options:
16508 Set mode of operation of the filter.
16510 Can be one of the following:
16514 If both @code{value} and @code{key} is set, select frames
16515 which have such metadata. If only @code{key} is set, select
16516 every frame that has such key in metadata.
16519 Add new metadata @code{key} and @code{value}. If key is already available
16523 Modify value of already present key.
16526 If @code{value} is set, delete only keys that have such value.
16527 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16531 Print key and its value if metadata was found. If @code{key} is not set print all
16532 metadata values available in frame.
16536 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16539 Set metadata value which will be used. This option is mandatory for
16540 @code{modify} and @code{add} mode.
16543 Which function to use when comparing metadata value and @code{value}.
16545 Can be one of following:
16549 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16552 Values are interpreted as strings, returns true if metadata value starts with
16553 the @code{value} option string.
16556 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16559 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16562 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16565 Values are interpreted as floats, returns true if expression from option @code{expr}
16570 Set expression which is used when @code{function} is set to @code{expr}.
16571 The expression is evaluated through the eval API and can contain the following
16576 Float representation of @code{value} from metadata key.
16579 Float representation of @code{value} as supplied by user in @code{value} option.
16583 If specified in @code{print} mode, output is written to the named file. Instead of
16584 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16585 for standard output. If @code{file} option is not set, output is written to the log
16586 with AV_LOG_INFO loglevel.
16590 @subsection Examples
16594 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16597 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16600 Print silencedetect output to file @file{metadata.txt}.
16602 silencedetect,ametadata=mode=print:file=metadata.txt
16605 Direct all metadata to a pipe with file descriptor 4.
16607 metadata=mode=print:file='pipe\:4'
16611 @section perms, aperms
16613 Set read/write permissions for the output frames.
16615 These filters are mainly aimed at developers to test direct path in the
16616 following filter in the filtergraph.
16618 The filters accept the following options:
16622 Select the permissions mode.
16624 It accepts the following values:
16627 Do nothing. This is the default.
16629 Set all the output frames read-only.
16631 Set all the output frames directly writable.
16633 Make the frame read-only if writable, and writable if read-only.
16635 Set each output frame read-only or writable randomly.
16639 Set the seed for the @var{random} mode, must be an integer included between
16640 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16641 @code{-1}, the filter will try to use a good random seed on a best effort
16645 Note: in case of auto-inserted filter between the permission filter and the
16646 following one, the permission might not be received as expected in that
16647 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16648 perms/aperms filter can avoid this problem.
16650 @section realtime, arealtime
16652 Slow down filtering to match real time approximatively.
16654 These filters will pause the filtering for a variable amount of time to
16655 match the output rate with the input timestamps.
16656 They are similar to the @option{re} option to @code{ffmpeg}.
16658 They accept the following options:
16662 Time limit for the pauses. Any pause longer than that will be considered
16663 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16667 @section select, aselect
16669 Select frames to pass in output.
16671 This filter accepts the following options:
16676 Set expression, which is evaluated for each input frame.
16678 If the expression is evaluated to zero, the frame is discarded.
16680 If the evaluation result is negative or NaN, the frame is sent to the
16681 first output; otherwise it is sent to the output with index
16682 @code{ceil(val)-1}, assuming that the input index starts from 0.
16684 For example a value of @code{1.2} corresponds to the output with index
16685 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16688 Set the number of outputs. The output to which to send the selected
16689 frame is based on the result of the evaluation. Default value is 1.
16692 The expression can contain the following constants:
16696 The (sequential) number of the filtered frame, starting from 0.
16699 The (sequential) number of the selected frame, starting from 0.
16701 @item prev_selected_n
16702 The sequential number of the last selected frame. It's NAN if undefined.
16705 The timebase of the input timestamps.
16708 The PTS (Presentation TimeStamp) of the filtered video frame,
16709 expressed in @var{TB} units. It's NAN if undefined.
16712 The PTS of the filtered video frame,
16713 expressed in seconds. It's NAN if undefined.
16716 The PTS of the previously filtered video frame. It's NAN if undefined.
16718 @item prev_selected_pts
16719 The PTS of the last previously filtered video frame. It's NAN if undefined.
16721 @item prev_selected_t
16722 The PTS of the last previously selected video frame. It's NAN if undefined.
16725 The PTS of the first video frame in the video. It's NAN if undefined.
16728 The time of the first video frame in the video. It's NAN if undefined.
16730 @item pict_type @emph{(video only)}
16731 The type of the filtered frame. It can assume one of the following
16743 @item interlace_type @emph{(video only)}
16744 The frame interlace type. It can assume one of the following values:
16747 The frame is progressive (not interlaced).
16749 The frame is top-field-first.
16751 The frame is bottom-field-first.
16754 @item consumed_sample_n @emph{(audio only)}
16755 the number of selected samples before the current frame
16757 @item samples_n @emph{(audio only)}
16758 the number of samples in the current frame
16760 @item sample_rate @emph{(audio only)}
16761 the input sample rate
16764 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16767 the position in the file of the filtered frame, -1 if the information
16768 is not available (e.g. for synthetic video)
16770 @item scene @emph{(video only)}
16771 value between 0 and 1 to indicate a new scene; a low value reflects a low
16772 probability for the current frame to introduce a new scene, while a higher
16773 value means the current frame is more likely to be one (see the example below)
16775 @item concatdec_select
16776 The concat demuxer can select only part of a concat input file by setting an
16777 inpoint and an outpoint, but the output packets may not be entirely contained
16778 in the selected interval. By using this variable, it is possible to skip frames
16779 generated by the concat demuxer which are not exactly contained in the selected
16782 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16783 and the @var{lavf.concat.duration} packet metadata values which are also
16784 present in the decoded frames.
16786 The @var{concatdec_select} variable is -1 if the frame pts is at least
16787 start_time and either the duration metadata is missing or the frame pts is less
16788 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16791 That basically means that an input frame is selected if its pts is within the
16792 interval set by the concat demuxer.
16796 The default value of the select expression is "1".
16798 @subsection Examples
16802 Select all frames in input:
16807 The example above is the same as:
16819 Select only I-frames:
16821 select='eq(pict_type\,I)'
16825 Select one frame every 100:
16827 select='not(mod(n\,100))'
16831 Select only frames contained in the 10-20 time interval:
16833 select=between(t\,10\,20)
16837 Select only I-frames contained in the 10-20 time interval:
16839 select=between(t\,10\,20)*eq(pict_type\,I)
16843 Select frames with a minimum distance of 10 seconds:
16845 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16849 Use aselect to select only audio frames with samples number > 100:
16851 aselect='gt(samples_n\,100)'
16855 Create a mosaic of the first scenes:
16857 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16860 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16864 Send even and odd frames to separate outputs, and compose them:
16866 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16870 Select useful frames from an ffconcat file which is using inpoints and
16871 outpoints but where the source files are not intra frame only.
16873 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16877 @section sendcmd, asendcmd
16879 Send commands to filters in the filtergraph.
16881 These filters read commands to be sent to other filters in the
16884 @code{sendcmd} must be inserted between two video filters,
16885 @code{asendcmd} must be inserted between two audio filters, but apart
16886 from that they act the same way.
16888 The specification of commands can be provided in the filter arguments
16889 with the @var{commands} option, or in a file specified by the
16890 @var{filename} option.
16892 These filters accept the following options:
16895 Set the commands to be read and sent to the other filters.
16897 Set the filename of the commands to be read and sent to the other
16901 @subsection Commands syntax
16903 A commands description consists of a sequence of interval
16904 specifications, comprising a list of commands to be executed when a
16905 particular event related to that interval occurs. The occurring event
16906 is typically the current frame time entering or leaving a given time
16909 An interval is specified by the following syntax:
16911 @var{START}[-@var{END}] @var{COMMANDS};
16914 The time interval is specified by the @var{START} and @var{END} times.
16915 @var{END} is optional and defaults to the maximum time.
16917 The current frame time is considered within the specified interval if
16918 it is included in the interval [@var{START}, @var{END}), that is when
16919 the time is greater or equal to @var{START} and is lesser than
16922 @var{COMMANDS} consists of a sequence of one or more command
16923 specifications, separated by ",", relating to that interval. The
16924 syntax of a command specification is given by:
16926 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16929 @var{FLAGS} is optional and specifies the type of events relating to
16930 the time interval which enable sending the specified command, and must
16931 be a non-null sequence of identifier flags separated by "+" or "|" and
16932 enclosed between "[" and "]".
16934 The following flags are recognized:
16937 The command is sent when the current frame timestamp enters the
16938 specified interval. In other words, the command is sent when the
16939 previous frame timestamp was not in the given interval, and the
16943 The command is sent when the current frame timestamp leaves the
16944 specified interval. In other words, the command is sent when the
16945 previous frame timestamp was in the given interval, and the
16949 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16952 @var{TARGET} specifies the target of the command, usually the name of
16953 the filter class or a specific filter instance name.
16955 @var{COMMAND} specifies the name of the command for the target filter.
16957 @var{ARG} is optional and specifies the optional list of argument for
16958 the given @var{COMMAND}.
16960 Between one interval specification and another, whitespaces, or
16961 sequences of characters starting with @code{#} until the end of line,
16962 are ignored and can be used to annotate comments.
16964 A simplified BNF description of the commands specification syntax
16967 @var{COMMAND_FLAG} ::= "enter" | "leave"
16968 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16969 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16970 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16971 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16972 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16975 @subsection Examples
16979 Specify audio tempo change at second 4:
16981 asendcmd=c='4.0 atempo tempo 1.5',atempo
16985 Specify a list of drawtext and hue commands in a file.
16987 # show text in the interval 5-10
16988 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16989 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16991 # desaturate the image in the interval 15-20
16992 15.0-20.0 [enter] hue s 0,
16993 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16995 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16997 # apply an exponential saturation fade-out effect, starting from time 25
16998 25 [enter] hue s exp(25-t)
17001 A filtergraph allowing to read and process the above command list
17002 stored in a file @file{test.cmd}, can be specified with:
17004 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
17009 @section setpts, asetpts
17011 Change the PTS (presentation timestamp) of the input frames.
17013 @code{setpts} works on video frames, @code{asetpts} on audio frames.
17015 This filter accepts the following options:
17020 The expression which is evaluated for each frame to construct its timestamp.
17024 The expression is evaluated through the eval API and can contain the following
17029 frame rate, only defined for constant frame-rate video
17032 The presentation timestamp in input
17035 The count of the input frame for video or the number of consumed samples,
17036 not including the current frame for audio, starting from 0.
17038 @item NB_CONSUMED_SAMPLES
17039 The number of consumed samples, not including the current frame (only
17042 @item NB_SAMPLES, S
17043 The number of samples in the current frame (only audio)
17045 @item SAMPLE_RATE, SR
17046 The audio sample rate.
17049 The PTS of the first frame.
17052 the time in seconds of the first frame
17055 State whether the current frame is interlaced.
17058 the time in seconds of the current frame
17061 original position in the file of the frame, or undefined if undefined
17062 for the current frame
17065 The previous input PTS.
17068 previous input time in seconds
17071 The previous output PTS.
17074 previous output time in seconds
17077 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
17081 The wallclock (RTC) time at the start of the movie in microseconds.
17084 The timebase of the input timestamps.
17088 @subsection Examples
17092 Start counting PTS from zero
17094 setpts=PTS-STARTPTS
17098 Apply fast motion effect:
17104 Apply slow motion effect:
17110 Set fixed rate of 25 frames per second:
17116 Set fixed rate 25 fps with some jitter:
17118 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17122 Apply an offset of 10 seconds to the input PTS:
17128 Generate timestamps from a "live source" and rebase onto the current timebase:
17130 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17134 Generate timestamps by counting samples:
17141 @section settb, asettb
17143 Set the timebase to use for the output frames timestamps.
17144 It is mainly useful for testing timebase configuration.
17146 It accepts the following parameters:
17151 The expression which is evaluated into the output timebase.
17155 The value for @option{tb} is an arithmetic expression representing a
17156 rational. The expression can contain the constants "AVTB" (the default
17157 timebase), "intb" (the input timebase) and "sr" (the sample rate,
17158 audio only). Default value is "intb".
17160 @subsection Examples
17164 Set the timebase to 1/25:
17170 Set the timebase to 1/10:
17176 Set the timebase to 1001/1000:
17182 Set the timebase to 2*intb:
17188 Set the default timebase value:
17195 Convert input audio to a video output representing frequency spectrum
17196 logarithmically using Brown-Puckette constant Q transform algorithm with
17197 direct frequency domain coefficient calculation (but the transform itself
17198 is not really constant Q, instead the Q factor is actually variable/clamped),
17199 with musical tone scale, from E0 to D#10.
17201 The filter accepts the following options:
17205 Specify the video size for the output. It must be even. For the syntax of this option,
17206 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17207 Default value is @code{1920x1080}.
17210 Set the output frame rate. Default value is @code{25}.
17213 Set the bargraph height. It must be even. Default value is @code{-1} which
17214 computes the bargraph height automatically.
17217 Set the axis height. It must be even. Default value is @code{-1} which computes
17218 the axis height automatically.
17221 Set the sonogram height. It must be even. Default value is @code{-1} which
17222 computes the sonogram height automatically.
17225 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17226 instead. Default value is @code{1}.
17228 @item sono_v, volume
17229 Specify the sonogram volume expression. It can contain variables:
17232 the @var{bar_v} evaluated expression
17233 @item frequency, freq, f
17234 the frequency where it is evaluated
17235 @item timeclamp, tc
17236 the value of @var{timeclamp} option
17240 @item a_weighting(f)
17241 A-weighting of equal loudness
17242 @item b_weighting(f)
17243 B-weighting of equal loudness
17244 @item c_weighting(f)
17245 C-weighting of equal loudness.
17247 Default value is @code{16}.
17249 @item bar_v, volume2
17250 Specify the bargraph volume expression. It can contain variables:
17253 the @var{sono_v} evaluated expression
17254 @item frequency, freq, f
17255 the frequency where it is evaluated
17256 @item timeclamp, tc
17257 the value of @var{timeclamp} option
17261 @item a_weighting(f)
17262 A-weighting of equal loudness
17263 @item b_weighting(f)
17264 B-weighting of equal loudness
17265 @item c_weighting(f)
17266 C-weighting of equal loudness.
17268 Default value is @code{sono_v}.
17270 @item sono_g, gamma
17271 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17272 higher gamma makes the spectrum having more range. Default value is @code{3}.
17273 Acceptable range is @code{[1, 7]}.
17275 @item bar_g, gamma2
17276 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17280 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17281 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17283 @item timeclamp, tc
17284 Specify the transform timeclamp. At low frequency, there is trade-off between
17285 accuracy in time domain and frequency domain. If timeclamp is lower,
17286 event in time domain is represented more accurately (such as fast bass drum),
17287 otherwise event in frequency domain is represented more accurately
17288 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17291 Set attack time in seconds. The default is @code{0} (disabled). Otherwise, it
17292 limits future samples by applying asymmetric windowing in time domain, useful
17293 when low latency is required. Accepted range is @code{[0, 1]}.
17296 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17297 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17300 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17301 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17304 This option is deprecated and ignored.
17307 Specify the transform length in time domain. Use this option to control accuracy
17308 trade-off between time domain and frequency domain at every frequency sample.
17309 It can contain variables:
17311 @item frequency, freq, f
17312 the frequency where it is evaluated
17313 @item timeclamp, tc
17314 the value of @var{timeclamp} option.
17316 Default value is @code{384*tc/(384+tc*f)}.
17319 Specify the transform count for every video frame. Default value is @code{6}.
17320 Acceptable range is @code{[1, 30]}.
17323 Specify the transform count for every single pixel. Default value is @code{0},
17324 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17327 Specify font file for use with freetype to draw the axis. If not specified,
17328 use embedded font. Note that drawing with font file or embedded font is not
17329 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17333 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17334 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17337 Specify font color expression. This is arithmetic expression that should return
17338 integer value 0xRRGGBB. It can contain variables:
17340 @item frequency, freq, f
17341 the frequency where it is evaluated
17342 @item timeclamp, tc
17343 the value of @var{timeclamp} option
17348 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17349 @item r(x), g(x), b(x)
17350 red, green, and blue value of intensity x.
17352 Default value is @code{st(0, (midi(f)-59.5)/12);
17353 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17354 r(1-ld(1)) + b(ld(1))}.
17357 Specify image file to draw the axis. This option override @var{fontfile} and
17358 @var{fontcolor} option.
17361 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17362 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17363 Default value is @code{1}.
17366 Set colorspace. The accepted values are:
17369 Unspecified (default)
17378 BT.470BG or BT.601-6 625
17381 SMPTE-170M or BT.601-6 525
17387 BT.2020 with non-constant luminance
17392 Set spectrogram color scheme. This is list of floating point values with format
17393 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17394 The default is @code{1|0.5|0|0|0.5|1}.
17398 @subsection Examples
17402 Playing audio while showing the spectrum:
17404 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17408 Same as above, but with frame rate 30 fps:
17410 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17414 Playing at 1280x720:
17416 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17420 Disable sonogram display:
17426 A1 and its harmonics: A1, A2, (near)E3, A3:
17428 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),
17429 asplit[a][out1]; [a] showcqt [out0]'
17433 Same as above, but with more accuracy in frequency domain:
17435 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),
17436 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17442 bar_v=10:sono_v=bar_v*a_weighting(f)
17446 Custom gamma, now spectrum is linear to the amplitude.
17452 Custom tlength equation:
17454 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)))'
17458 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17460 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17464 Custom font using fontconfig:
17466 font='Courier New,Monospace,mono|bold'
17470 Custom frequency range with custom axis using image file:
17472 axisfile=myaxis.png:basefreq=40:endfreq=10000
17478 Convert input audio to video output representing the audio power spectrum.
17479 Audio amplitude is on Y-axis while frequency is on X-axis.
17481 The filter accepts the following options:
17485 Specify size of video. For the syntax of this option, check the
17486 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17487 Default is @code{1024x512}.
17491 This set how each frequency bin will be represented.
17493 It accepts the following values:
17499 Default is @code{bar}.
17502 Set amplitude scale.
17504 It accepts the following values:
17518 Default is @code{log}.
17521 Set frequency scale.
17523 It accepts the following values:
17532 Reverse logarithmic scale.
17534 Default is @code{lin}.
17539 It accepts the following values:
17555 Default is @code{w2048}
17558 Set windowing function.
17560 It accepts the following values:
17582 Default is @code{hanning}.
17585 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17586 which means optimal overlap for selected window function will be picked.
17589 Set time averaging. Setting this to 0 will display current maximal peaks.
17590 Default is @code{1}, which means time averaging is disabled.
17593 Specify list of colors separated by space or by '|' which will be used to
17594 draw channel frequencies. Unrecognized or missing colors will be replaced
17598 Set channel display mode.
17600 It accepts the following values:
17605 Default is @code{combined}.
17608 Set minimum amplitude used in @code{log} amplitude scaler.
17612 @anchor{showspectrum}
17613 @section showspectrum
17615 Convert input audio to a video output, representing the audio frequency
17618 The filter accepts the following options:
17622 Specify the video size for the output. For the syntax of this option, check the
17623 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17624 Default value is @code{640x512}.
17627 Specify how the spectrum should slide along the window.
17629 It accepts the following values:
17632 the samples start again on the left when they reach the right
17634 the samples scroll from right to left
17636 frames are only produced when the samples reach the right
17638 the samples scroll from left to right
17641 Default value is @code{replace}.
17644 Specify display mode.
17646 It accepts the following values:
17649 all channels are displayed in the same row
17651 all channels are displayed in separate rows
17654 Default value is @samp{combined}.
17657 Specify display color mode.
17659 It accepts the following values:
17662 each channel is displayed in a separate color
17664 each channel is displayed using the same color scheme
17666 each channel is displayed using the rainbow color scheme
17668 each channel is displayed using the moreland color scheme
17670 each channel is displayed using the nebulae color scheme
17672 each channel is displayed using the fire color scheme
17674 each channel is displayed using the fiery color scheme
17676 each channel is displayed using the fruit color scheme
17678 each channel is displayed using the cool color scheme
17681 Default value is @samp{channel}.
17684 Specify scale used for calculating intensity color values.
17686 It accepts the following values:
17691 square root, default
17702 Default value is @samp{sqrt}.
17705 Set saturation modifier for displayed colors. Negative values provide
17706 alternative color scheme. @code{0} is no saturation at all.
17707 Saturation must be in [-10.0, 10.0] range.
17708 Default value is @code{1}.
17711 Set window function.
17713 It accepts the following values:
17737 Default value is @code{hann}.
17740 Set orientation of time vs frequency axis. Can be @code{vertical} or
17741 @code{horizontal}. Default is @code{vertical}.
17744 Set ratio of overlap window. Default value is @code{0}.
17745 When value is @code{1} overlap is set to recommended size for specific
17746 window function currently used.
17749 Set scale gain for calculating intensity color values.
17750 Default value is @code{1}.
17753 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17756 Set color rotation, must be in [-1.0, 1.0] range.
17757 Default value is @code{0}.
17760 The usage is very similar to the showwaves filter; see the examples in that
17763 @subsection Examples
17767 Large window with logarithmic color scaling:
17769 showspectrum=s=1280x480:scale=log
17773 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17775 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17776 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17780 @section showspectrumpic
17782 Convert input audio to a single video frame, representing the audio frequency
17785 The filter accepts the following options:
17789 Specify the video size for the output. For the syntax of this option, check the
17790 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17791 Default value is @code{4096x2048}.
17794 Specify display mode.
17796 It accepts the following values:
17799 all channels are displayed in the same row
17801 all channels are displayed in separate rows
17803 Default value is @samp{combined}.
17806 Specify display color mode.
17808 It accepts the following values:
17811 each channel is displayed in a separate color
17813 each channel is displayed using the same color scheme
17815 each channel is displayed using the rainbow color scheme
17817 each channel is displayed using the moreland color scheme
17819 each channel is displayed using the nebulae color scheme
17821 each channel is displayed using the fire color scheme
17823 each channel is displayed using the fiery color scheme
17825 each channel is displayed using the fruit color scheme
17827 each channel is displayed using the cool color scheme
17829 Default value is @samp{intensity}.
17832 Specify scale used for calculating intensity color values.
17834 It accepts the following values:
17839 square root, default
17849 Default value is @samp{log}.
17852 Set saturation modifier for displayed colors. Negative values provide
17853 alternative color scheme. @code{0} is no saturation at all.
17854 Saturation must be in [-10.0, 10.0] range.
17855 Default value is @code{1}.
17858 Set window function.
17860 It accepts the following values:
17883 Default value is @code{hann}.
17886 Set orientation of time vs frequency axis. Can be @code{vertical} or
17887 @code{horizontal}. Default is @code{vertical}.
17890 Set scale gain for calculating intensity color values.
17891 Default value is @code{1}.
17894 Draw time and frequency axes and legends. Default is enabled.
17897 Set color rotation, must be in [-1.0, 1.0] range.
17898 Default value is @code{0}.
17901 @subsection Examples
17905 Extract an audio spectrogram of a whole audio track
17906 in a 1024x1024 picture using @command{ffmpeg}:
17908 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17912 @section showvolume
17914 Convert input audio volume to a video output.
17916 The filter accepts the following options:
17923 Set border width, allowed range is [0, 5]. Default is 1.
17926 Set channel width, allowed range is [80, 8192]. Default is 400.
17929 Set channel height, allowed range is [1, 900]. Default is 20.
17932 Set fade, allowed range is [0.001, 1]. Default is 0.95.
17935 Set volume color expression.
17937 The expression can use the following variables:
17941 Current max volume of channel in dB.
17947 Current channel number, starting from 0.
17951 If set, displays channel names. Default is enabled.
17954 If set, displays volume values. Default is enabled.
17957 Set orientation, can be @code{horizontal} or @code{vertical},
17958 default is @code{horizontal}.
17961 Set step size, allowed range s [0, 5]. Default is 0, which means
17967 Convert input audio to a video output, representing the samples waves.
17969 The filter accepts the following options:
17973 Specify the video size for the output. For the syntax of this option, check the
17974 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17975 Default value is @code{600x240}.
17980 Available values are:
17983 Draw a point for each sample.
17986 Draw a vertical line for each sample.
17989 Draw a point for each sample and a line between them.
17992 Draw a centered vertical line for each sample.
17995 Default value is @code{point}.
17998 Set the number of samples which are printed on the same column. A
17999 larger value will decrease the frame rate. Must be a positive
18000 integer. This option can be set only if the value for @var{rate}
18001 is not explicitly specified.
18004 Set the (approximate) output frame rate. This is done by setting the
18005 option @var{n}. Default value is "25".
18007 @item split_channels
18008 Set if channels should be drawn separately or overlap. Default value is 0.
18011 Set colors separated by '|' which are going to be used for drawing of each channel.
18014 Set amplitude scale.
18016 Available values are:
18034 @subsection Examples
18038 Output the input file audio and the corresponding video representation
18041 amovie=a.mp3,asplit[out0],showwaves[out1]
18045 Create a synthetic signal and show it with showwaves, forcing a
18046 frame rate of 30 frames per second:
18048 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
18052 @section showwavespic
18054 Convert input audio to a single video frame, representing the samples waves.
18056 The filter accepts the following options:
18060 Specify the video size for the output. For the syntax of this option, check the
18061 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
18062 Default value is @code{600x240}.
18064 @item split_channels
18065 Set if channels should be drawn separately or overlap. Default value is 0.
18068 Set colors separated by '|' which are going to be used for drawing of each channel.
18071 Set amplitude scale.
18073 Available values are:
18091 @subsection Examples
18095 Extract a channel split representation of the wave form of a whole audio track
18096 in a 1024x800 picture using @command{ffmpeg}:
18098 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
18102 @section sidedata, asidedata
18104 Delete frame side data, or select frames based on it.
18106 This filter accepts the following options:
18110 Set mode of operation of the filter.
18112 Can be one of the following:
18116 Select every frame with side data of @code{type}.
18119 Delete side data of @code{type}. If @code{type} is not set, delete all side
18125 Set side data type used with all modes. Must be set for @code{select} mode. For
18126 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18127 in @file{libavutil/frame.h}. For example, to choose
18128 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18132 @section spectrumsynth
18134 Sythesize audio from 2 input video spectrums, first input stream represents
18135 magnitude across time and second represents phase across time.
18136 The filter will transform from frequency domain as displayed in videos back
18137 to time domain as presented in audio output.
18139 This filter is primarily created for reversing processed @ref{showspectrum}
18140 filter outputs, but can synthesize sound from other spectrograms too.
18141 But in such case results are going to be poor if the phase data is not
18142 available, because in such cases phase data need to be recreated, usually
18143 its just recreated from random noise.
18144 For best results use gray only output (@code{channel} color mode in
18145 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
18146 @code{lin} scale for phase video. To produce phase, for 2nd video, use
18147 @code{data} option. Inputs videos should generally use @code{fullframe}
18148 slide mode as that saves resources needed for decoding video.
18150 The filter accepts the following options:
18154 Specify sample rate of output audio, the sample rate of audio from which
18155 spectrum was generated may differ.
18158 Set number of channels represented in input video spectrums.
18161 Set scale which was used when generating magnitude input spectrum.
18162 Can be @code{lin} or @code{log}. Default is @code{log}.
18165 Set slide which was used when generating inputs spectrums.
18166 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18167 Default is @code{fullframe}.
18170 Set window function used for resynthesis.
18173 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18174 which means optimal overlap for selected window function will be picked.
18177 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18178 Default is @code{vertical}.
18181 @subsection Examples
18185 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18186 then resynthesize videos back to audio with spectrumsynth:
18188 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
18189 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
18190 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18194 @section split, asplit
18196 Split input into several identical outputs.
18198 @code{asplit} works with audio input, @code{split} with video.
18200 The filter accepts a single parameter which specifies the number of outputs. If
18201 unspecified, it defaults to 2.
18203 @subsection Examples
18207 Create two separate outputs from the same input:
18209 [in] split [out0][out1]
18213 To create 3 or more outputs, you need to specify the number of
18216 [in] asplit=3 [out0][out1][out2]
18220 Create two separate outputs from the same input, one cropped and
18223 [in] split [splitout1][splitout2];
18224 [splitout1] crop=100:100:0:0 [cropout];
18225 [splitout2] pad=200:200:100:100 [padout];
18229 Create 5 copies of the input audio with @command{ffmpeg}:
18231 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18237 Receive commands sent through a libzmq client, and forward them to
18238 filters in the filtergraph.
18240 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18241 must be inserted between two video filters, @code{azmq} between two
18244 To enable these filters you need to install the libzmq library and
18245 headers and configure FFmpeg with @code{--enable-libzmq}.
18247 For more information about libzmq see:
18248 @url{http://www.zeromq.org/}
18250 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18251 receives messages sent through a network interface defined by the
18252 @option{bind_address} option.
18254 The received message must be in the form:
18256 @var{TARGET} @var{COMMAND} [@var{ARG}]
18259 @var{TARGET} specifies the target of the command, usually the name of
18260 the filter class or a specific filter instance name.
18262 @var{COMMAND} specifies the name of the command for the target filter.
18264 @var{ARG} is optional and specifies the optional argument list for the
18265 given @var{COMMAND}.
18267 Upon reception, the message is processed and the corresponding command
18268 is injected into the filtergraph. Depending on the result, the filter
18269 will send a reply to the client, adopting the format:
18271 @var{ERROR_CODE} @var{ERROR_REASON}
18275 @var{MESSAGE} is optional.
18277 @subsection Examples
18279 Look at @file{tools/zmqsend} for an example of a zmq client which can
18280 be used to send commands processed by these filters.
18282 Consider the following filtergraph generated by @command{ffplay}
18284 ffplay -dumpgraph 1 -f lavfi "
18285 color=s=100x100:c=red [l];
18286 color=s=100x100:c=blue [r];
18287 nullsrc=s=200x100, zmq [bg];
18288 [bg][l] overlay [bg+l];
18289 [bg+l][r] overlay=x=100 "
18292 To change the color of the left side of the video, the following
18293 command can be used:
18295 echo Parsed_color_0 c yellow | tools/zmqsend
18298 To change the right side:
18300 echo Parsed_color_1 c pink | tools/zmqsend
18303 @c man end MULTIMEDIA FILTERS
18305 @chapter Multimedia Sources
18306 @c man begin MULTIMEDIA SOURCES
18308 Below is a description of the currently available multimedia sources.
18312 This is the same as @ref{movie} source, except it selects an audio
18318 Read audio and/or video stream(s) from a movie container.
18320 It accepts the following parameters:
18324 The name of the resource to read (not necessarily a file; it can also be a
18325 device or a stream accessed through some protocol).
18327 @item format_name, f
18328 Specifies the format assumed for the movie to read, and can be either
18329 the name of a container or an input device. If not specified, the
18330 format is guessed from @var{movie_name} or by probing.
18332 @item seek_point, sp
18333 Specifies the seek point in seconds. The frames will be output
18334 starting from this seek point. The parameter is evaluated with
18335 @code{av_strtod}, so the numerical value may be suffixed by an IS
18336 postfix. The default value is "0".
18339 Specifies the streams to read. Several streams can be specified,
18340 separated by "+". The source will then have as many outputs, in the
18341 same order. The syntax is explained in the ``Stream specifiers''
18342 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18343 respectively the default (best suited) video and audio stream. Default
18344 is "dv", or "da" if the filter is called as "amovie".
18346 @item stream_index, si
18347 Specifies the index of the video stream to read. If the value is -1,
18348 the most suitable video stream will be automatically selected. The default
18349 value is "-1". Deprecated. If the filter is called "amovie", it will select
18350 audio instead of video.
18353 Specifies how many times to read the stream in sequence.
18354 If the value is 0, the stream will be looped infinitely.
18355 Default value is "1".
18357 Note that when the movie is looped the source timestamps are not
18358 changed, so it will generate non monotonically increasing timestamps.
18360 @item discontinuity
18361 Specifies the time difference between frames above which the point is
18362 considered a timestamp discontinuity which is removed by adjusting the later
18366 It allows overlaying a second video on top of the main input of
18367 a filtergraph, as shown in this graph:
18369 input -----------> deltapts0 --> overlay --> output
18372 movie --> scale--> deltapts1 -------+
18374 @subsection Examples
18378 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18379 on top of the input labelled "in":
18381 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18382 [in] setpts=PTS-STARTPTS [main];
18383 [main][over] overlay=16:16 [out]
18387 Read from a video4linux2 device, and overlay it on top of the input
18390 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18391 [in] setpts=PTS-STARTPTS [main];
18392 [main][over] overlay=16:16 [out]
18396 Read the first video stream and the audio stream with id 0x81 from
18397 dvd.vob; the video is connected to the pad named "video" and the audio is
18398 connected to the pad named "audio":
18400 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18404 @subsection Commands
18406 Both movie and amovie support the following commands:
18409 Perform seek using "av_seek_frame".
18410 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18413 @var{stream_index}: If stream_index is -1, a default
18414 stream is selected, and @var{timestamp} is automatically converted
18415 from AV_TIME_BASE units to the stream specific time_base.
18417 @var{timestamp}: Timestamp in AVStream.time_base units
18418 or, if no stream is specified, in AV_TIME_BASE units.
18420 @var{flags}: Flags which select direction and seeking mode.
18424 Get movie duration in AV_TIME_BASE units.
18428 @c man end MULTIMEDIA SOURCES