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.
1049 The filter accepts the following options:
1053 Set the number of loops.
1056 Set maximal number of samples.
1059 Set first sample of loop.
1065 Merge two or more audio streams into a single multi-channel stream.
1067 The filter accepts the following options:
1072 Set the number of inputs. Default is 2.
1076 If the channel layouts of the inputs are disjoint, and therefore compatible,
1077 the channel layout of the output will be set accordingly and the channels
1078 will be reordered as necessary. If the channel layouts of the inputs are not
1079 disjoint, the output will have all the channels of the first input then all
1080 the channels of the second input, in that order, and the channel layout of
1081 the output will be the default value corresponding to the total number of
1084 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1085 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1086 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1087 first input, b1 is the first channel of the second input).
1089 On the other hand, if both input are in stereo, the output channels will be
1090 in the default order: a1, a2, b1, b2, and the channel layout will be
1091 arbitrarily set to 4.0, which may or may not be the expected value.
1093 All inputs must have the same sample rate, and format.
1095 If inputs do not have the same duration, the output will stop with the
1098 @subsection Examples
1102 Merge two mono files into a stereo stream:
1104 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1108 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1110 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
1116 Mixes multiple audio inputs into a single output.
1118 Note that this filter only supports float samples (the @var{amerge}
1119 and @var{pan} audio filters support many formats). If the @var{amix}
1120 input has integer samples then @ref{aresample} will be automatically
1121 inserted to perform the conversion to float samples.
1125 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1127 will mix 3 input audio streams to a single output with the same duration as the
1128 first input and a dropout transition time of 3 seconds.
1130 It accepts the following parameters:
1134 The number of inputs. If unspecified, it defaults to 2.
1137 How to determine the end-of-stream.
1141 The duration of the longest input. (default)
1144 The duration of the shortest input.
1147 The duration of the first input.
1151 @item dropout_transition
1152 The transition time, in seconds, for volume renormalization when an input
1153 stream ends. The default value is 2 seconds.
1157 @section anequalizer
1159 High-order parametric multiband equalizer for each channel.
1161 It accepts the following parameters:
1165 This option string is in format:
1166 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1167 Each equalizer band is separated by '|'.
1171 Set channel number to which equalization will be applied.
1172 If input doesn't have that channel the entry is ignored.
1175 Set central frequency for band.
1176 If input doesn't have that frequency the entry is ignored.
1179 Set band width in hertz.
1182 Set band gain in dB.
1185 Set filter type for band, optional, can be:
1189 Butterworth, this is default.
1200 With this option activated frequency response of anequalizer is displayed
1204 Set video stream size. Only useful if curves option is activated.
1207 Set max gain that will be displayed. Only useful if curves option is activated.
1208 Setting this to a reasonable value makes it possible to display gain which is derived from
1209 neighbour bands which are too close to each other and thus produce higher gain
1210 when both are activated.
1213 Set frequency scale used to draw frequency response in video output.
1214 Can be linear or logarithmic. Default is logarithmic.
1217 Set color for each channel curve which is going to be displayed in video stream.
1218 This is list of color names separated by space or by '|'.
1219 Unrecognised or missing colors will be replaced by white color.
1222 @subsection Examples
1226 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1227 for first 2 channels using Chebyshev type 1 filter:
1229 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1233 @subsection Commands
1235 This filter supports the following commands:
1238 Alter existing filter parameters.
1239 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1241 @var{fN} is existing filter number, starting from 0, if no such filter is available
1243 @var{freq} set new frequency parameter.
1244 @var{width} set new width parameter in herz.
1245 @var{gain} set new gain parameter in dB.
1247 Full filter invocation with asendcmd may look like this:
1248 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1253 Pass the audio source unchanged to the output.
1257 Pad the end of an audio stream with silence.
1259 This can be used together with @command{ffmpeg} @option{-shortest} to
1260 extend audio streams to the same length as the video stream.
1262 A description of the accepted options follows.
1266 Set silence packet size. Default value is 4096.
1269 Set the number of samples of silence to add to the end. After the
1270 value is reached, the stream is terminated. This option is mutually
1271 exclusive with @option{whole_len}.
1274 Set the minimum total number of samples in the output audio stream. If
1275 the value is longer than the input audio length, silence is added to
1276 the end, until the value is reached. This option is mutually exclusive
1277 with @option{pad_len}.
1280 If neither the @option{pad_len} nor the @option{whole_len} option is
1281 set, the filter will add silence to the end of the input stream
1284 @subsection Examples
1288 Add 1024 samples of silence to the end of the input:
1294 Make sure the audio output will contain at least 10000 samples, pad
1295 the input with silence if required:
1297 apad=whole_len=10000
1301 Use @command{ffmpeg} to pad the audio input with silence, so that the
1302 video stream will always result the shortest and will be converted
1303 until the end in the output file when using the @option{shortest}
1306 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1311 Add a phasing effect to the input audio.
1313 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1314 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1316 A description of the accepted parameters follows.
1320 Set input gain. Default is 0.4.
1323 Set output gain. Default is 0.74
1326 Set delay in milliseconds. Default is 3.0.
1329 Set decay. Default is 0.4.
1332 Set modulation speed in Hz. Default is 0.5.
1335 Set modulation type. Default is triangular.
1337 It accepts the following values:
1346 Audio pulsator is something between an autopanner and a tremolo.
1347 But it can produce funny stereo effects as well. Pulsator changes the volume
1348 of the left and right channel based on a LFO (low frequency oscillator) with
1349 different waveforms and shifted phases.
1350 This filter have the ability to define an offset between left and right
1351 channel. An offset of 0 means that both LFO shapes match each other.
1352 The left and right channel are altered equally - a conventional tremolo.
1353 An offset of 50% means that the shape of the right channel is exactly shifted
1354 in phase (or moved backwards about half of the frequency) - pulsator acts as
1355 an autopanner. At 1 both curves match again. Every setting in between moves the
1356 phase shift gapless between all stages and produces some "bypassing" sounds with
1357 sine and triangle waveforms. The more you set the offset near 1 (starting from
1358 the 0.5) the faster the signal passes from the left to the right speaker.
1360 The filter accepts the following options:
1364 Set input gain. By default it is 1. Range is [0.015625 - 64].
1367 Set output gain. By default it is 1. Range is [0.015625 - 64].
1370 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1371 sawup or sawdown. Default is sine.
1374 Set modulation. Define how much of original signal is affected by the LFO.
1377 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1380 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1383 Set pulse width. Default is 1. Allowed range is [0 - 2].
1386 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1389 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1393 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1397 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1398 if timing is set to hz.
1404 Resample the input audio to the specified parameters, using the
1405 libswresample library. If none are specified then the filter will
1406 automatically convert between its input and output.
1408 This filter is also able to stretch/squeeze the audio data to make it match
1409 the timestamps or to inject silence / cut out audio to make it match the
1410 timestamps, do a combination of both or do neither.
1412 The filter accepts the syntax
1413 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1414 expresses a sample rate and @var{resampler_options} is a list of
1415 @var{key}=@var{value} pairs, separated by ":". See the
1416 @ref{Resampler Options,,the "Resampler Options" section in the
1417 ffmpeg-resampler(1) manual,ffmpeg-resampler}
1418 for the complete list of supported options.
1420 @subsection Examples
1424 Resample the input audio to 44100Hz:
1430 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1431 samples per second compensation:
1433 aresample=async=1000
1439 Reverse an audio clip.
1441 Warning: This filter requires memory to buffer the entire clip, so trimming
1444 @subsection Examples
1448 Take the first 5 seconds of a clip, and reverse it.
1450 atrim=end=5,areverse
1454 @section asetnsamples
1456 Set the number of samples per each output audio frame.
1458 The last output packet may contain a different number of samples, as
1459 the filter will flush all the remaining samples when the input audio
1462 The filter accepts the following options:
1466 @item nb_out_samples, n
1467 Set the number of frames per each output audio frame. The number is
1468 intended as the number of samples @emph{per each channel}.
1469 Default value is 1024.
1472 If set to 1, the filter will pad the last audio frame with zeroes, so
1473 that the last frame will contain the same number of samples as the
1474 previous ones. Default value is 1.
1477 For example, to set the number of per-frame samples to 1234 and
1478 disable padding for the last frame, use:
1480 asetnsamples=n=1234:p=0
1485 Set the sample rate without altering the PCM data.
1486 This will result in a change of speed and pitch.
1488 The filter accepts the following options:
1491 @item sample_rate, r
1492 Set the output sample rate. Default is 44100 Hz.
1497 Show a line containing various information for each input audio frame.
1498 The input audio is not modified.
1500 The shown line contains a sequence of key/value pairs of the form
1501 @var{key}:@var{value}.
1503 The following values are shown in the output:
1507 The (sequential) number of the input frame, starting from 0.
1510 The presentation timestamp of the input frame, in time base units; the time base
1511 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1514 The presentation timestamp of the input frame in seconds.
1517 position of the frame in the input stream, -1 if this information in
1518 unavailable and/or meaningless (for example in case of synthetic audio)
1527 The sample rate for the audio frame.
1530 The number of samples (per channel) in the frame.
1533 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1534 audio, the data is treated as if all the planes were concatenated.
1536 @item plane_checksums
1537 A list of Adler-32 checksums for each data plane.
1543 Display time domain statistical information about the audio channels.
1544 Statistics are calculated and displayed for each audio channel and,
1545 where applicable, an overall figure is also given.
1547 It accepts the following option:
1550 Short window length in seconds, used for peak and trough RMS measurement.
1551 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1555 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1556 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1559 Available keys for each channel are:
1590 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1591 this @code{lavfi.astats.Overall.Peak_count}.
1593 For description what each key means read below.
1596 Set number of frame after which stats are going to be recalculated.
1597 Default is disabled.
1600 A description of each shown parameter follows:
1604 Mean amplitude displacement from zero.
1607 Minimal sample level.
1610 Maximal sample level.
1612 @item Min difference
1613 Minimal difference between two consecutive samples.
1615 @item Max difference
1616 Maximal difference between two consecutive samples.
1618 @item Mean difference
1619 Mean difference between two consecutive samples.
1620 The average of each difference between two consecutive samples.
1624 Standard peak and RMS level measured in dBFS.
1628 Peak and trough values for RMS level measured over a short window.
1631 Standard ratio of peak to RMS level (note: not in dB).
1634 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1635 (i.e. either @var{Min level} or @var{Max level}).
1638 Number of occasions (not the number of samples) that the signal attained either
1639 @var{Min level} or @var{Max level}.
1642 Overall bit depth of audio. Number of bits used for each sample.
1649 The filter accepts exactly one parameter, the audio tempo. If not
1650 specified then the filter will assume nominal 1.0 tempo. Tempo must
1651 be in the [0.5, 2.0] range.
1653 @subsection Examples
1657 Slow down audio to 80% tempo:
1663 To speed up audio to 125% tempo:
1671 Trim the input so that the output contains one continuous subpart of the input.
1673 It accepts the following parameters:
1676 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1677 sample with the timestamp @var{start} will be the first sample in the output.
1680 Specify time of the first audio sample that will be dropped, i.e. the
1681 audio sample immediately preceding the one with the timestamp @var{end} will be
1682 the last sample in the output.
1685 Same as @var{start}, except this option sets the start timestamp in samples
1689 Same as @var{end}, except this option sets the end timestamp in samples instead
1693 The maximum duration of the output in seconds.
1696 The number of the first sample that should be output.
1699 The number of the first sample that should be dropped.
1702 @option{start}, @option{end}, and @option{duration} are expressed as time
1703 duration specifications; see
1704 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1706 Note that the first two sets of the start/end options and the @option{duration}
1707 option look at the frame timestamp, while the _sample options simply count the
1708 samples that pass through the filter. So start/end_pts and start/end_sample will
1709 give different results when the timestamps are wrong, inexact or do not start at
1710 zero. Also note that this filter does not modify the timestamps. If you wish
1711 to have the output timestamps start at zero, insert the asetpts filter after the
1714 If multiple start or end options are set, this filter tries to be greedy and
1715 keep all samples that match at least one of the specified constraints. To keep
1716 only the part that matches all the constraints at once, chain multiple atrim
1719 The defaults are such that all the input is kept. So it is possible to set e.g.
1720 just the end values to keep everything before the specified time.
1725 Drop everything except the second minute of input:
1727 ffmpeg -i INPUT -af atrim=60:120
1731 Keep only the first 1000 samples:
1733 ffmpeg -i INPUT -af atrim=end_sample=1000
1740 Apply a two-pole Butterworth band-pass filter with central
1741 frequency @var{frequency}, and (3dB-point) band-width width.
1742 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1743 instead of the default: constant 0dB peak gain.
1744 The filter roll off at 6dB per octave (20dB per decade).
1746 The filter accepts the following options:
1750 Set the filter's central frequency. Default is @code{3000}.
1753 Constant skirt gain if set to 1. Defaults to 0.
1756 Set method to specify band-width of filter.
1769 Specify the band-width of a filter in width_type units.
1774 Apply a two-pole Butterworth band-reject filter with central
1775 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1776 The filter roll off at 6dB per octave (20dB per decade).
1778 The filter accepts the following options:
1782 Set the filter's central frequency. Default is @code{3000}.
1785 Set method to specify band-width of filter.
1798 Specify the band-width of a filter in width_type units.
1803 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1804 shelving filter with a response similar to that of a standard
1805 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1807 The filter accepts the following options:
1811 Give the gain at 0 Hz. Its useful range is about -20
1812 (for a large cut) to +20 (for a large boost).
1813 Beware of clipping when using a positive gain.
1816 Set the filter's central frequency and so can be used
1817 to extend or reduce the frequency range to be boosted or cut.
1818 The default value is @code{100} Hz.
1821 Set method to specify band-width of filter.
1834 Determine how steep is the filter's shelf transition.
1839 Apply a biquad IIR filter with the given coefficients.
1840 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1841 are the numerator and denominator coefficients respectively.
1844 Bauer stereo to binaural transformation, which improves headphone listening of
1845 stereo audio records.
1847 It accepts the following parameters:
1851 Pre-defined crossfeed level.
1855 Default level (fcut=700, feed=50).
1858 Chu Moy circuit (fcut=700, feed=60).
1861 Jan Meier circuit (fcut=650, feed=95).
1866 Cut frequency (in Hz).
1875 Remap input channels to new locations.
1877 It accepts the following parameters:
1880 Map channels from input to output. The argument is a '|'-separated list of
1881 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1882 @var{in_channel} form. @var{in_channel} can be either the name of the input
1883 channel (e.g. FL for front left) or its index in the input channel layout.
1884 @var{out_channel} is the name of the output channel or its index in the output
1885 channel layout. If @var{out_channel} is not given then it is implicitly an
1886 index, starting with zero and increasing by one for each mapping.
1888 @item channel_layout
1889 The channel layout of the output stream.
1892 If no mapping is present, the filter will implicitly map input channels to
1893 output channels, preserving indices.
1895 For example, assuming a 5.1+downmix input MOV file,
1897 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1899 will create an output WAV file tagged as stereo from the downmix channels of
1902 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1904 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1907 @section channelsplit
1909 Split each channel from an input audio stream into a separate output stream.
1911 It accepts the following parameters:
1913 @item channel_layout
1914 The channel layout of the input stream. The default is "stereo".
1917 For example, assuming a stereo input MP3 file,
1919 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1921 will create an output Matroska file with two audio streams, one containing only
1922 the left channel and the other the right channel.
1924 Split a 5.1 WAV file into per-channel files:
1926 ffmpeg -i in.wav -filter_complex
1927 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1928 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1929 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1934 Add a chorus effect to the audio.
1936 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1938 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1939 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1940 The modulation depth defines the range the modulated delay is played before or after
1941 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1942 sound tuned around the original one, like in a chorus where some vocals are slightly
1945 It accepts the following parameters:
1948 Set input gain. Default is 0.4.
1951 Set output gain. Default is 0.4.
1954 Set delays. A typical delay is around 40ms to 60ms.
1966 @subsection Examples
1972 chorus=0.7:0.9:55:0.4:0.25:2
1978 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1982 Fuller sounding chorus with three delays:
1984 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
1989 Compress or expand the audio's dynamic range.
1991 It accepts the following parameters:
1997 A list of times in seconds for each channel over which the instantaneous level
1998 of the input signal is averaged to determine its volume. @var{attacks} refers to
1999 increase of volume and @var{decays} refers to decrease of volume. For most
2000 situations, the attack time (response to the audio getting louder) should be
2001 shorter than the decay time, because the human ear is more sensitive to sudden
2002 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
2003 a typical value for decay is 0.8 seconds.
2004 If specified number of attacks & decays is lower than number of channels, the last
2005 set attack/decay will be used for all remaining channels.
2008 A list of points for the transfer function, specified in dB relative to the
2009 maximum possible signal amplitude. Each key points list must be defined using
2010 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
2011 @code{x0/y0 x1/y1 x2/y2 ....}
2013 The input values must be in strictly increasing order but the transfer function
2014 does not have to be monotonically rising. The point @code{0/0} is assumed but
2015 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
2016 function are @code{-70/-70|-60/-20}.
2019 Set the curve radius in dB for all joints. It defaults to 0.01.
2022 Set the additional gain in dB to be applied at all points on the transfer
2023 function. This allows for easy adjustment of the overall gain.
2027 Set an initial volume, in dB, to be assumed for each channel when filtering
2028 starts. This permits the user to supply a nominal level initially, so that, for
2029 example, a very large gain is not applied to initial signal levels before the
2030 companding has begun to operate. A typical value for audio which is initially
2031 quiet is -90 dB. It defaults to 0.
2034 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
2035 delayed before being fed to the volume adjuster. Specifying a delay
2036 approximately equal to the attack/decay times allows the filter to effectively
2037 operate in predictive rather than reactive mode. It defaults to 0.
2041 @subsection Examples
2045 Make music with both quiet and loud passages suitable for listening to in a
2048 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2051 Another example for audio with whisper and explosion parts:
2053 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2057 A noise gate for when the noise is at a lower level than the signal:
2059 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2063 Here is another noise gate, this time for when the noise is at a higher level
2064 than the signal (making it, in some ways, similar to squelch):
2066 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2070 2:1 compression starting at -6dB:
2072 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2076 2:1 compression starting at -9dB:
2078 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2082 2:1 compression starting at -12dB:
2084 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2088 2:1 compression starting at -18dB:
2090 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2094 3:1 compression starting at -15dB:
2096 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2102 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2108 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
2112 Hard limiter at -6dB:
2114 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2118 Hard limiter at -12dB:
2120 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2124 Hard noise gate at -35 dB:
2126 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2132 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2136 @section compensationdelay
2138 Compensation Delay Line is a metric based delay to compensate differing
2139 positions of microphones or speakers.
2141 For example, you have recorded guitar with two microphones placed in
2142 different location. Because the front of sound wave has fixed speed in
2143 normal conditions, the phasing of microphones can vary and depends on
2144 their location and interposition. The best sound mix can be achieved when
2145 these microphones are in phase (synchronized). Note that distance of
2146 ~30 cm between microphones makes one microphone to capture signal in
2147 antiphase to another microphone. That makes the final mix sounding moody.
2148 This filter helps to solve phasing problems by adding different delays
2149 to each microphone track and make them synchronized.
2151 The best result can be reached when you take one track as base and
2152 synchronize other tracks one by one with it.
2153 Remember that synchronization/delay tolerance depends on sample rate, too.
2154 Higher sample rates will give more tolerance.
2156 It accepts the following parameters:
2160 Set millimeters distance. This is compensation distance for fine tuning.
2164 Set cm distance. This is compensation distance for tightening distance setup.
2168 Set meters distance. This is compensation distance for hard distance setup.
2172 Set dry amount. Amount of unprocessed (dry) signal.
2176 Set wet amount. Amount of processed (wet) signal.
2180 Set temperature degree in Celsius. This is the temperature of the environment.
2184 @section crystalizer
2185 Simple algorithm to expand audio dynamic range.
2187 The filter accepts the following options:
2191 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2192 (unchanged sound) to 10.0 (maximum effect).
2195 Enable clipping. By default is enabled.
2199 Apply a DC shift to the audio.
2201 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2202 in the recording chain) from the audio. The effect of a DC offset is reduced
2203 headroom and hence volume. The @ref{astats} filter can be used to determine if
2204 a signal has a DC offset.
2208 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2212 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2213 used to prevent clipping.
2217 Dynamic Audio Normalizer.
2219 This filter applies a certain amount of gain to the input audio in order
2220 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2221 contrast to more "simple" normalization algorithms, the Dynamic Audio
2222 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2223 This allows for applying extra gain to the "quiet" sections of the audio
2224 while avoiding distortions or clipping the "loud" sections. In other words:
2225 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2226 sections, in the sense that the volume of each section is brought to the
2227 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2228 this goal *without* applying "dynamic range compressing". It will retain 100%
2229 of the dynamic range *within* each section of the audio file.
2233 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2234 Default is 500 milliseconds.
2235 The Dynamic Audio Normalizer processes the input audio in small chunks,
2236 referred to as frames. This is required, because a peak magnitude has no
2237 meaning for just a single sample value. Instead, we need to determine the
2238 peak magnitude for a contiguous sequence of sample values. While a "standard"
2239 normalizer would simply use the peak magnitude of the complete file, the
2240 Dynamic Audio Normalizer determines the peak magnitude individually for each
2241 frame. The length of a frame is specified in milliseconds. By default, the
2242 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2243 been found to give good results with most files.
2244 Note that the exact frame length, in number of samples, will be determined
2245 automatically, based on the sampling rate of the individual input audio file.
2248 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2249 number. Default is 31.
2250 Probably the most important parameter of the Dynamic Audio Normalizer is the
2251 @code{window size} of the Gaussian smoothing filter. The filter's window size
2252 is specified in frames, centered around the current frame. For the sake of
2253 simplicity, this must be an odd number. Consequently, the default value of 31
2254 takes into account the current frame, as well as the 15 preceding frames and
2255 the 15 subsequent frames. Using a larger window results in a stronger
2256 smoothing effect and thus in less gain variation, i.e. slower gain
2257 adaptation. Conversely, using a smaller window results in a weaker smoothing
2258 effect and thus in more gain variation, i.e. faster gain adaptation.
2259 In other words, the more you increase this value, the more the Dynamic Audio
2260 Normalizer will behave like a "traditional" normalization filter. On the
2261 contrary, the more you decrease this value, the more the Dynamic Audio
2262 Normalizer will behave like a dynamic range compressor.
2265 Set the target peak value. This specifies the highest permissible magnitude
2266 level for the normalized audio input. This filter will try to approach the
2267 target peak magnitude as closely as possible, but at the same time it also
2268 makes sure that the normalized signal will never exceed the peak magnitude.
2269 A frame's maximum local gain factor is imposed directly by the target peak
2270 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2271 It is not recommended to go above this value.
2274 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2275 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2276 factor for each input frame, i.e. the maximum gain factor that does not
2277 result in clipping or distortion. The maximum gain factor is determined by
2278 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2279 additionally bounds the frame's maximum gain factor by a predetermined
2280 (global) maximum gain factor. This is done in order to avoid excessive gain
2281 factors in "silent" or almost silent frames. By default, the maximum gain
2282 factor is 10.0, For most inputs the default value should be sufficient and
2283 it usually is not recommended to increase this value. Though, for input
2284 with an extremely low overall volume level, it may be necessary to allow even
2285 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2286 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2287 Instead, a "sigmoid" threshold function will be applied. This way, the
2288 gain factors will smoothly approach the threshold value, but never exceed that
2292 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2293 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2294 This means that the maximum local gain factor for each frame is defined
2295 (only) by the frame's highest magnitude sample. This way, the samples can
2296 be amplified as much as possible without exceeding the maximum signal
2297 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2298 Normalizer can also take into account the frame's root mean square,
2299 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2300 determine the power of a time-varying signal. It is therefore considered
2301 that the RMS is a better approximation of the "perceived loudness" than
2302 just looking at the signal's peak magnitude. Consequently, by adjusting all
2303 frames to a constant RMS value, a uniform "perceived loudness" can be
2304 established. If a target RMS value has been specified, a frame's local gain
2305 factor is defined as the factor that would result in exactly that RMS value.
2306 Note, however, that the maximum local gain factor is still restricted by the
2307 frame's highest magnitude sample, in order to prevent clipping.
2310 Enable channels coupling. By default is enabled.
2311 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2312 amount. This means the same gain factor will be applied to all channels, i.e.
2313 the maximum possible gain factor is determined by the "loudest" channel.
2314 However, in some recordings, it may happen that the volume of the different
2315 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2316 In this case, this option can be used to disable the channel coupling. This way,
2317 the gain factor will be determined independently for each channel, depending
2318 only on the individual channel's highest magnitude sample. This allows for
2319 harmonizing the volume of the different channels.
2322 Enable DC bias correction. By default is disabled.
2323 An audio signal (in the time domain) is a sequence of sample values.
2324 In the Dynamic Audio Normalizer these sample values are represented in the
2325 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2326 audio signal, or "waveform", should be centered around the zero point.
2327 That means if we calculate the mean value of all samples in a file, or in a
2328 single frame, then the result should be 0.0 or at least very close to that
2329 value. If, however, there is a significant deviation of the mean value from
2330 0.0, in either positive or negative direction, this is referred to as a
2331 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2332 Audio Normalizer provides optional DC bias correction.
2333 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2334 the mean value, or "DC correction" offset, of each input frame and subtract
2335 that value from all of the frame's sample values which ensures those samples
2336 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2337 boundaries, the DC correction offset values will be interpolated smoothly
2338 between neighbouring frames.
2341 Enable alternative boundary mode. By default is disabled.
2342 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2343 around each frame. This includes the preceding frames as well as the
2344 subsequent frames. However, for the "boundary" frames, located at the very
2345 beginning and at the very end of the audio file, not all neighbouring
2346 frames are available. In particular, for the first few frames in the audio
2347 file, the preceding frames are not known. And, similarly, for the last few
2348 frames in the audio file, the subsequent frames are not known. Thus, the
2349 question arises which gain factors should be assumed for the missing frames
2350 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2351 to deal with this situation. The default boundary mode assumes a gain factor
2352 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2353 "fade out" at the beginning and at the end of the input, respectively.
2356 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2357 By default, the Dynamic Audio Normalizer does not apply "traditional"
2358 compression. This means that signal peaks will not be pruned and thus the
2359 full dynamic range will be retained within each local neighbourhood. However,
2360 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2361 normalization algorithm with a more "traditional" compression.
2362 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2363 (thresholding) function. If (and only if) the compression feature is enabled,
2364 all input frames will be processed by a soft knee thresholding function prior
2365 to the actual normalization process. Put simply, the thresholding function is
2366 going to prune all samples whose magnitude exceeds a certain threshold value.
2367 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2368 value. Instead, the threshold value will be adjusted for each individual
2370 In general, smaller parameters result in stronger compression, and vice versa.
2371 Values below 3.0 are not recommended, because audible distortion may appear.
2376 Make audio easier to listen to on headphones.
2378 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2379 so that when listened to on headphones the stereo image is moved from
2380 inside your head (standard for headphones) to outside and in front of
2381 the listener (standard for speakers).
2387 Apply a two-pole peaking equalisation (EQ) filter. With this
2388 filter, the signal-level at and around a selected frequency can
2389 be increased or decreased, whilst (unlike bandpass and bandreject
2390 filters) that at all other frequencies is unchanged.
2392 In order to produce complex equalisation curves, this filter can
2393 be given several times, each with a different central frequency.
2395 The filter accepts the following options:
2399 Set the filter's central frequency in Hz.
2402 Set method to specify band-width of filter.
2415 Specify the band-width of a filter in width_type units.
2418 Set the required gain or attenuation in dB.
2419 Beware of clipping when using a positive gain.
2422 @subsection Examples
2425 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2427 equalizer=f=1000:width_type=h:width=200:g=-10
2431 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2433 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2437 @section extrastereo
2439 Linearly increases the difference between left and right channels which
2440 adds some sort of "live" effect to playback.
2442 The filter accepts the following options:
2446 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2447 (average of both channels), with 1.0 sound will be unchanged, with
2448 -1.0 left and right channels will be swapped.
2451 Enable clipping. By default is enabled.
2454 @section firequalizer
2455 Apply FIR Equalization using arbitrary frequency response.
2457 The filter accepts the following option:
2461 Set gain curve equation (in dB). The expression can contain variables:
2464 the evaluated frequency
2468 channel number, set to 0 when multichannels evaluation is disabled
2470 channel id, see libavutil/channel_layout.h, set to the first channel id when
2471 multichannels evaluation is disabled
2475 channel_layout, see libavutil/channel_layout.h
2480 @item gain_interpolate(f)
2481 interpolate gain on frequency f based on gain_entry
2482 @item cubic_interpolate(f)
2483 same as gain_interpolate, but smoother
2485 This option is also available as command. Default is @code{gain_interpolate(f)}.
2488 Set gain entry for gain_interpolate function. The expression can
2492 store gain entry at frequency f with value g
2494 This option is also available as command.
2497 Set filter delay in seconds. Higher value means more accurate.
2498 Default is @code{0.01}.
2501 Set filter accuracy in Hz. Lower value means more accurate.
2502 Default is @code{5}.
2505 Set window function. Acceptable values are:
2508 rectangular window, useful when gain curve is already smooth
2510 hann window (default)
2516 3-terms continuous 1st derivative nuttall window
2518 minimum 3-terms discontinuous nuttall window
2520 4-terms continuous 1st derivative nuttall window
2522 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2524 blackman-harris window
2530 If enabled, use fixed number of audio samples. This improves speed when
2531 filtering with large delay. Default is disabled.
2534 Enable multichannels evaluation on gain. Default is disabled.
2537 Enable zero phase mode by subtracting timestamp to compensate delay.
2538 Default is disabled.
2541 Set scale used by gain. Acceptable values are:
2544 linear frequency, linear gain
2546 linear frequency, logarithmic (in dB) gain (default)
2548 logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2550 logarithmic frequency, logarithmic gain
2554 Set file for dumping, suitable for gnuplot.
2557 Set scale for dumpfile. Acceptable values are same with scale option.
2561 Enable 2-channel convolution using complex FFT. This improves speed significantly.
2562 Default is disabled.
2565 @subsection Examples
2570 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2573 lowpass at 1000 Hz with gain_entry:
2575 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2578 custom equalization:
2580 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2583 higher delay with zero phase to compensate delay:
2585 firequalizer=delay=0.1:fixed=on:zero_phase=on
2588 lowpass on left channel, highpass on right channel:
2590 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2591 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2596 Apply a flanging effect to the audio.
2598 The filter accepts the following options:
2602 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2605 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2608 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2612 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2613 Default value is 71.
2616 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2619 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2620 Default value is @var{sinusoidal}.
2623 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2624 Default value is 25.
2627 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2628 Default is @var{linear}.
2633 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
2634 embedded HDCD codes is expanded into a 20-bit PCM stream.
2636 The filter supports the Peak Extend and Low-level Gain Adjustment features
2637 of HDCD, and detects the Transient Filter flag.
2640 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
2643 When using the filter with wav, note the default encoding for wav is 16-bit,
2644 so the resulting 20-bit stream will be truncated back to 16-bit. Use something
2645 like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
2647 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2648 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2651 The filter accepts the following options:
2654 @item disable_autoconvert
2655 Disable any automatic format conversion or resampling in the filter graph.
2657 @item process_stereo
2658 Process the stereo channels together. If target_gain does not match between
2659 channels, consider it invalid and use the last valid target_gain.
2662 Set the code detect timer period in ms.
2665 Always extend peaks above -3dBFS even if PE isn't signaled.
2668 Replace audio with a solid tone and adjust the amplitude to signal some
2669 specific aspect of the decoding process. The output file can be loaded in
2670 an audio editor alongside the original to aid analysis.
2672 @code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2679 Gain adjustment level at each sample
2681 Samples where peak extend occurs
2683 Samples where the code detect timer is active
2685 Samples where the target gain does not match between channels
2691 Apply a high-pass filter with 3dB point frequency.
2692 The filter can be either single-pole, or double-pole (the default).
2693 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2695 The filter accepts the following options:
2699 Set frequency in Hz. Default is 3000.
2702 Set number of poles. Default is 2.
2705 Set method to specify band-width of filter.
2718 Specify the band-width of a filter in width_type units.
2719 Applies only to double-pole filter.
2720 The default is 0.707q and gives a Butterworth response.
2725 Join multiple input streams into one multi-channel stream.
2727 It accepts the following parameters:
2731 The number of input streams. It defaults to 2.
2733 @item channel_layout
2734 The desired output channel layout. It defaults to stereo.
2737 Map channels from inputs to output. The argument is a '|'-separated list of
2738 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2739 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2740 can be either the name of the input channel (e.g. FL for front left) or its
2741 index in the specified input stream. @var{out_channel} is the name of the output
2745 The filter will attempt to guess the mappings when they are not specified
2746 explicitly. It does so by first trying to find an unused matching input channel
2747 and if that fails it picks the first unused input channel.
2749 Join 3 inputs (with properly set channel layouts):
2751 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2754 Build a 5.1 output from 6 single-channel streams:
2756 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2757 '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'
2763 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2765 To enable compilation of this filter you need to configure FFmpeg with
2766 @code{--enable-ladspa}.
2770 Specifies the name of LADSPA plugin library to load. If the environment
2771 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2772 each one of the directories specified by the colon separated list in
2773 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2774 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2775 @file{/usr/lib/ladspa/}.
2778 Specifies the plugin within the library. Some libraries contain only
2779 one plugin, but others contain many of them. If this is not set filter
2780 will list all available plugins within the specified library.
2783 Set the '|' separated list of controls which are zero or more floating point
2784 values that determine the behavior of the loaded plugin (for example delay,
2786 Controls need to be defined using the following syntax:
2787 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2788 @var{valuei} is the value set on the @var{i}-th control.
2789 Alternatively they can be also defined using the following syntax:
2790 @var{value0}|@var{value1}|@var{value2}|..., where
2791 @var{valuei} is the value set on the @var{i}-th control.
2792 If @option{controls} is set to @code{help}, all available controls and
2793 their valid ranges are printed.
2795 @item sample_rate, s
2796 Specify the sample rate, default to 44100. Only used if plugin have
2800 Set the number of samples per channel per each output frame, default
2801 is 1024. Only used if plugin have zero inputs.
2804 Set the minimum duration of the sourced audio. See
2805 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2806 for the accepted syntax.
2807 Note that the resulting duration may be greater than the specified duration,
2808 as the generated audio is always cut at the end of a complete frame.
2809 If not specified, or the expressed duration is negative, the audio is
2810 supposed to be generated forever.
2811 Only used if plugin have zero inputs.
2815 @subsection Examples
2819 List all available plugins within amp (LADSPA example plugin) library:
2825 List all available controls and their valid ranges for @code{vcf_notch}
2826 plugin from @code{VCF} library:
2828 ladspa=f=vcf:p=vcf_notch:c=help
2832 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2835 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2839 Add reverberation to the audio using TAP-plugins
2840 (Tom's Audio Processing plugins):
2842 ladspa=file=tap_reverb:tap_reverb
2846 Generate white noise, with 0.2 amplitude:
2848 ladspa=file=cmt:noise_source_white:c=c0=.2
2852 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2853 @code{C* Audio Plugin Suite} (CAPS) library:
2855 ladspa=file=caps:Click:c=c1=20'
2859 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2861 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2865 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2866 @code{SWH Plugins} collection:
2868 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2872 Attenuate low frequencies using Multiband EQ from Steve Harris
2873 @code{SWH Plugins} collection:
2875 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2879 @subsection Commands
2881 This filter supports the following commands:
2884 Modify the @var{N}-th control value.
2886 If the specified value is not valid, it is ignored and prior one is kept.
2891 EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2892 Support for both single pass (livestreams, files) and double pass (files) modes.
2893 This algorithm can target IL, LRA, and maximum true peak.
2895 The filter accepts the following options:
2899 Set integrated loudness target.
2900 Range is -70.0 - -5.0. Default value is -24.0.
2903 Set loudness range target.
2904 Range is 1.0 - 20.0. Default value is 7.0.
2907 Set maximum true peak.
2908 Range is -9.0 - +0.0. Default value is -2.0.
2910 @item measured_I, measured_i
2911 Measured IL of input file.
2912 Range is -99.0 - +0.0.
2914 @item measured_LRA, measured_lra
2915 Measured LRA of input file.
2916 Range is 0.0 - 99.0.
2918 @item measured_TP, measured_tp
2919 Measured true peak of input file.
2920 Range is -99.0 - +99.0.
2922 @item measured_thresh
2923 Measured threshold of input file.
2924 Range is -99.0 - +0.0.
2927 Set offset gain. Gain is applied before the true-peak limiter.
2928 Range is -99.0 - +99.0. Default is +0.0.
2931 Normalize linearly if possible.
2932 measured_I, measured_LRA, measured_TP, and measured_thresh must also
2933 to be specified in order to use this mode.
2934 Options are true or false. Default is true.
2937 Treat mono input files as "dual-mono". If a mono file is intended for playback
2938 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
2939 If set to @code{true}, this option will compensate for this effect.
2940 Multi-channel input files are not affected by this option.
2941 Options are true or false. Default is false.
2944 Set print format for stats. Options are summary, json, or none.
2945 Default value is none.
2950 Apply a low-pass filter with 3dB point frequency.
2951 The filter can be either single-pole or double-pole (the default).
2952 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2954 The filter accepts the following options:
2958 Set frequency in Hz. Default is 500.
2961 Set number of poles. Default is 2.
2964 Set method to specify band-width of filter.
2977 Specify the band-width of a filter in width_type units.
2978 Applies only to double-pole filter.
2979 The default is 0.707q and gives a Butterworth response.
2985 Mix channels with specific gain levels. The filter accepts the output
2986 channel layout followed by a set of channels definitions.
2988 This filter is also designed to efficiently remap the channels of an audio
2991 The filter accepts parameters of the form:
2992 "@var{l}|@var{outdef}|@var{outdef}|..."
2996 output channel layout or number of channels
2999 output channel specification, of the form:
3000 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3003 output channel to define, either a channel name (FL, FR, etc.) or a channel
3004 number (c0, c1, etc.)
3007 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3010 input channel to use, see out_name for details; it is not possible to mix
3011 named and numbered input channels
3014 If the `=' in a channel specification is replaced by `<', then the gains for
3015 that specification will be renormalized so that the total is 1, thus
3016 avoiding clipping noise.
3018 @subsection Mixing examples
3020 For example, if you want to down-mix from stereo to mono, but with a bigger
3021 factor for the left channel:
3023 pan=1c|c0=0.9*c0+0.1*c1
3026 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3027 7-channels surround:
3029 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3032 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3033 that should be preferred (see "-ac" option) unless you have very specific
3036 @subsection Remapping examples
3038 The channel remapping will be effective if, and only if:
3041 @item gain coefficients are zeroes or ones,
3042 @item only one input per channel output,
3045 If all these conditions are satisfied, the filter will notify the user ("Pure
3046 channel mapping detected"), and use an optimized and lossless method to do the
3049 For example, if you have a 5.1 source and want a stereo audio stream by
3050 dropping the extra channels:
3052 pan="stereo| c0=FL | c1=FR"
3055 Given the same source, you can also switch front left and front right channels
3056 and keep the input channel layout:
3058 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3061 If the input is a stereo audio stream, you can mute the front left channel (and
3062 still keep the stereo channel layout) with:
3067 Still with a stereo audio stream input, you can copy the right channel in both
3068 front left and right:
3070 pan="stereo| c0=FR | c1=FR"
3075 ReplayGain scanner filter. This filter takes an audio stream as an input and
3076 outputs it unchanged.
3077 At end of filtering it displays @code{track_gain} and @code{track_peak}.
3081 Convert the audio sample format, sample rate and channel layout. It is
3082 not meant to be used directly.
3085 Apply time-stretching and pitch-shifting with librubberband.
3087 The filter accepts the following options:
3091 Set tempo scale factor.
3094 Set pitch scale factor.
3097 Set transients detector.
3098 Possible values are:
3107 Possible values are:
3116 Possible values are:
3123 Set processing window size.
3124 Possible values are:
3133 Possible values are:
3140 Enable formant preservation when shift pitching.
3141 Possible values are:
3149 Possible values are:
3158 Possible values are:
3165 @section sidechaincompress
3167 This filter acts like normal compressor but has the ability to compress
3168 detected signal using second input signal.
3169 It needs two input streams and returns one output stream.
3170 First input stream will be processed depending on second stream signal.
3171 The filtered signal then can be filtered with other filters in later stages of
3172 processing. See @ref{pan} and @ref{amerge} filter.
3174 The filter accepts the following options:
3178 Set input gain. Default is 1. Range is between 0.015625 and 64.
3181 If a signal of second stream raises above this level it will affect the gain
3182 reduction of first stream.
3183 By default is 0.125. Range is between 0.00097563 and 1.
3186 Set a ratio about which the signal is reduced. 1:2 means that if the level
3187 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3188 Default is 2. Range is between 1 and 20.
3191 Amount of milliseconds the signal has to rise above the threshold before gain
3192 reduction starts. Default is 20. Range is between 0.01 and 2000.
3195 Amount of milliseconds the signal has to fall below the threshold before
3196 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3199 Set the amount by how much signal will be amplified after processing.
3200 Default is 2. Range is from 1 and 64.
3203 Curve the sharp knee around the threshold to enter gain reduction more softly.
3204 Default is 2.82843. Range is between 1 and 8.
3207 Choose if the @code{average} level between all channels of side-chain stream
3208 or the louder(@code{maximum}) channel of side-chain stream affects the
3209 reduction. Default is @code{average}.
3212 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3213 of @code{rms}. Default is @code{rms} which is mainly smoother.
3216 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3219 How much to use compressed signal in output. Default is 1.
3220 Range is between 0 and 1.
3223 @subsection Examples
3227 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3228 depending on the signal of 2nd input and later compressed signal to be
3229 merged with 2nd input:
3231 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3235 @section sidechaingate
3237 A sidechain gate acts like a normal (wideband) gate but has the ability to
3238 filter the detected signal before sending it to the gain reduction stage.
3239 Normally a gate uses the full range signal to detect a level above the
3241 For example: If you cut all lower frequencies from your sidechain signal
3242 the gate will decrease the volume of your track only if not enough highs
3243 appear. With this technique you are able to reduce the resonation of a
3244 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3246 It needs two input streams and returns one output stream.
3247 First input stream will be processed depending on second stream signal.
3249 The filter accepts the following options:
3253 Set input level before filtering.
3254 Default is 1. Allowed range is from 0.015625 to 64.
3257 Set the level of gain reduction when the signal is below the threshold.
3258 Default is 0.06125. Allowed range is from 0 to 1.
3261 If a signal rises above this level the gain reduction is released.
3262 Default is 0.125. Allowed range is from 0 to 1.
3265 Set a ratio about which the signal is reduced.
3266 Default is 2. Allowed range is from 1 to 9000.
3269 Amount of milliseconds the signal has to rise above the threshold before gain
3271 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3274 Amount of milliseconds the signal has to fall below the threshold before the
3275 reduction is increased again. Default is 250 milliseconds.
3276 Allowed range is from 0.01 to 9000.
3279 Set amount of amplification of signal after processing.
3280 Default is 1. Allowed range is from 1 to 64.
3283 Curve the sharp knee around the threshold to enter gain reduction more softly.
3284 Default is 2.828427125. Allowed range is from 1 to 8.
3287 Choose if exact signal should be taken for detection or an RMS like one.
3288 Default is rms. Can be peak or rms.
3291 Choose if the average level between all channels or the louder channel affects
3293 Default is average. Can be average or maximum.
3296 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3299 @section silencedetect
3301 Detect silence in an audio stream.
3303 This filter logs a message when it detects that the input audio volume is less
3304 or equal to a noise tolerance value for a duration greater or equal to the
3305 minimum detected noise duration.
3307 The printed times and duration are expressed in seconds.
3309 The filter accepts the following options:
3313 Set silence duration until notification (default is 2 seconds).
3316 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3317 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3320 @subsection Examples
3324 Detect 5 seconds of silence with -50dB noise tolerance:
3326 silencedetect=n=-50dB:d=5
3330 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3331 tolerance in @file{silence.mp3}:
3333 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3337 @section silenceremove
3339 Remove silence from the beginning, middle or end of the audio.
3341 The filter accepts the following options:
3345 This value is used to indicate if audio should be trimmed at beginning of
3346 the audio. A value of zero indicates no silence should be trimmed from the
3347 beginning. When specifying a non-zero value, it trims audio up until it
3348 finds non-silence. Normally, when trimming silence from beginning of audio
3349 the @var{start_periods} will be @code{1} but it can be increased to higher
3350 values to trim all audio up to specific count of non-silence periods.
3351 Default value is @code{0}.
3353 @item start_duration
3354 Specify the amount of time that non-silence must be detected before it stops
3355 trimming audio. By increasing the duration, bursts of noises can be treated
3356 as silence and trimmed off. Default value is @code{0}.
3358 @item start_threshold
3359 This indicates what sample value should be treated as silence. For digital
3360 audio, a value of @code{0} may be fine but for audio recorded from analog,
3361 you may wish to increase the value to account for background noise.
3362 Can be specified in dB (in case "dB" is appended to the specified value)
3363 or amplitude ratio. Default value is @code{0}.
3366 Set the count for trimming silence from the end of audio.
3367 To remove silence from the middle of a file, specify a @var{stop_periods}
3368 that is negative. This value is then treated as a positive value and is
3369 used to indicate the effect should restart processing as specified by
3370 @var{start_periods}, making it suitable for removing periods of silence
3371 in the middle of the audio.
3372 Default value is @code{0}.
3375 Specify a duration of silence that must exist before audio is not copied any
3376 more. By specifying a higher duration, silence that is wanted can be left in
3378 Default value is @code{0}.
3380 @item stop_threshold
3381 This is the same as @option{start_threshold} but for trimming silence from
3383 Can be specified in dB (in case "dB" is appended to the specified value)
3384 or amplitude ratio. Default value is @code{0}.
3387 This indicates that @var{stop_duration} length of audio should be left intact
3388 at the beginning of each period of silence.
3389 For example, if you want to remove long pauses between words but do not want
3390 to remove the pauses completely. Default value is @code{0}.
3393 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3394 and works better with digital silence which is exactly 0.
3395 Default value is @code{rms}.
3398 Set ratio used to calculate size of window for detecting silence.
3399 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3402 @subsection Examples
3406 The following example shows how this filter can be used to start a recording
3407 that does not contain the delay at the start which usually occurs between
3408 pressing the record button and the start of the performance:
3410 silenceremove=1:5:0.02
3414 Trim all silence encountered from beginning to end where there is more than 1
3415 second of silence in audio:
3417 silenceremove=0:0:0:-1:1:-90dB
3423 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3424 loudspeakers around the user for binaural listening via headphones (audio
3425 formats up to 9 channels supported).
3426 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3427 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3428 Austrian Academy of Sciences.
3430 To enable compilation of this filter you need to configure FFmpeg with
3431 @code{--enable-netcdf}.
3433 The filter accepts the following options:
3437 Set the SOFA file used for rendering.
3440 Set gain applied to audio. Value is in dB. Default is 0.
3443 Set rotation of virtual loudspeakers in deg. Default is 0.
3446 Set elevation of virtual speakers in deg. Default is 0.
3449 Set distance in meters between loudspeakers and the listener with near-field
3450 HRTFs. Default is 1.
3453 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3454 processing audio in time domain which is slow.
3455 @var{freq} is processing audio in frequency domain which is fast.
3456 Default is @var{freq}.
3459 Set custom positions of virtual loudspeakers. Syntax for this option is:
3460 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3461 Each virtual loudspeaker is described with short channel name following with
3462 azimuth and elevation in degreees.
3463 Each virtual loudspeaker description is separated by '|'.
3464 For example to override front left and front right channel positions use:
3465 'speakers=FL 45 15|FR 345 15'.
3466 Descriptions with unrecognised channel names are ignored.
3469 @subsection Examples
3473 Using ClubFritz6 sofa file:
3475 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3479 Using ClubFritz12 sofa file and bigger radius with small rotation:
3481 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3485 Similar as above but with custom speaker positions for front left, front right, back left and back right
3486 and also with custom gain:
3488 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3492 @section stereotools
3494 This filter has some handy utilities to manage stereo signals, for converting
3495 M/S stereo recordings to L/R signal while having control over the parameters
3496 or spreading the stereo image of master track.
3498 The filter accepts the following options:
3502 Set input level before filtering for both channels. Defaults is 1.
3503 Allowed range is from 0.015625 to 64.
3506 Set output level after filtering for both channels. Defaults is 1.
3507 Allowed range is from 0.015625 to 64.
3510 Set input balance between both channels. Default is 0.
3511 Allowed range is from -1 to 1.
3514 Set output balance between both channels. Default is 0.
3515 Allowed range is from -1 to 1.
3518 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3519 clipping. Disabled by default.
3522 Mute the left channel. Disabled by default.
3525 Mute the right channel. Disabled by default.
3528 Change the phase of the left channel. Disabled by default.
3531 Change the phase of the right channel. Disabled by default.
3534 Set stereo mode. Available values are:
3538 Left/Right to Left/Right, this is default.
3541 Left/Right to Mid/Side.
3544 Mid/Side to Left/Right.
3547 Left/Right to Left/Left.
3550 Left/Right to Right/Right.
3553 Left/Right to Left + Right.
3556 Left/Right to Right/Left.
3560 Set level of side signal. Default is 1.
3561 Allowed range is from 0.015625 to 64.
3564 Set balance of side signal. Default is 0.
3565 Allowed range is from -1 to 1.
3568 Set level of the middle signal. Default is 1.
3569 Allowed range is from 0.015625 to 64.
3572 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3575 Set stereo base between mono and inversed channels. Default is 0.
3576 Allowed range is from -1 to 1.
3579 Set delay in milliseconds how much to delay left from right channel and
3580 vice versa. Default is 0. Allowed range is from -20 to 20.
3583 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3586 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3589 @subsection Examples
3593 Apply karaoke like effect:
3595 stereotools=mlev=0.015625
3599 Convert M/S signal to L/R:
3601 "stereotools=mode=ms>lr"
3605 @section stereowiden
3607 This filter enhance the stereo effect by suppressing signal common to both
3608 channels and by delaying the signal of left into right and vice versa,
3609 thereby widening the stereo effect.
3611 The filter accepts the following options:
3615 Time in milliseconds of the delay of left signal into right and vice versa.
3616 Default is 20 milliseconds.
3619 Amount of gain in delayed signal into right and vice versa. Gives a delay
3620 effect of left signal in right output and vice versa which gives widening
3621 effect. Default is 0.3.
3624 Cross feed of left into right with inverted phase. This helps in suppressing
3625 the mono. If the value is 1 it will cancel all the signal common to both
3626 channels. Default is 0.3.
3629 Set level of input signal of original channel. Default is 0.8.
3634 Boost or cut treble (upper) frequencies of the audio using a two-pole
3635 shelving filter with a response similar to that of a standard
3636 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3638 The filter accepts the following options:
3642 Give the gain at whichever is the lower of ~22 kHz and the
3643 Nyquist frequency. Its useful range is about -20 (for a large cut)
3644 to +20 (for a large boost). Beware of clipping when using a positive gain.
3647 Set the filter's central frequency and so can be used
3648 to extend or reduce the frequency range to be boosted or cut.
3649 The default value is @code{3000} Hz.
3652 Set method to specify band-width of filter.
3665 Determine how steep is the filter's shelf transition.
3670 Sinusoidal amplitude modulation.
3672 The filter accepts the following options:
3676 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3677 (20 Hz or lower) will result in a tremolo effect.
3678 This filter may also be used as a ring modulator by specifying
3679 a modulation frequency higher than 20 Hz.
3680 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3683 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3684 Default value is 0.5.
3689 Sinusoidal phase modulation.
3691 The filter accepts the following options:
3695 Modulation frequency in Hertz.
3696 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3699 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3700 Default value is 0.5.
3705 Adjust the input audio volume.
3707 It accepts the following parameters:
3711 Set audio volume expression.
3713 Output values are clipped to the maximum value.
3715 The output audio volume is given by the relation:
3717 @var{output_volume} = @var{volume} * @var{input_volume}
3720 The default value for @var{volume} is "1.0".
3723 This parameter represents the mathematical precision.
3725 It determines which input sample formats will be allowed, which affects the
3726 precision of the volume scaling.
3730 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3732 32-bit floating-point; this limits input sample format to FLT. (default)
3734 64-bit floating-point; this limits input sample format to DBL.
3738 Choose the behaviour on encountering ReplayGain side data in input frames.
3742 Remove ReplayGain side data, ignoring its contents (the default).
3745 Ignore ReplayGain side data, but leave it in the frame.
3748 Prefer the track gain, if present.
3751 Prefer the album gain, if present.
3754 @item replaygain_preamp
3755 Pre-amplification gain in dB to apply to the selected replaygain gain.
3757 Default value for @var{replaygain_preamp} is 0.0.
3760 Set when the volume expression is evaluated.
3762 It accepts the following values:
3765 only evaluate expression once during the filter initialization, or
3766 when the @samp{volume} command is sent
3769 evaluate expression for each incoming frame
3772 Default value is @samp{once}.
3775 The volume expression can contain the following parameters.
3779 frame number (starting at zero)
3782 @item nb_consumed_samples
3783 number of samples consumed by the filter
3785 number of samples in the current frame
3787 original frame position in the file
3793 PTS at start of stream
3795 time at start of stream
3801 last set volume value
3804 Note that when @option{eval} is set to @samp{once} only the
3805 @var{sample_rate} and @var{tb} variables are available, all other
3806 variables will evaluate to NAN.
3808 @subsection Commands
3810 This filter supports the following commands:
3813 Modify the volume expression.
3814 The command accepts the same syntax of the corresponding option.
3816 If the specified expression is not valid, it is kept at its current
3818 @item replaygain_noclip
3819 Prevent clipping by limiting the gain applied.
3821 Default value for @var{replaygain_noclip} is 1.
3825 @subsection Examples
3829 Halve the input audio volume:
3833 volume=volume=-6.0206dB
3836 In all the above example the named key for @option{volume} can be
3837 omitted, for example like in:
3843 Increase input audio power by 6 decibels using fixed-point precision:
3845 volume=volume=6dB:precision=fixed
3849 Fade volume after time 10 with an annihilation period of 5 seconds:
3851 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3855 @section volumedetect
3857 Detect the volume of the input video.
3859 The filter has no parameters. The input is not modified. Statistics about
3860 the volume will be printed in the log when the input stream end is reached.
3862 In particular it will show the mean volume (root mean square), maximum
3863 volume (on a per-sample basis), and the beginning of a histogram of the
3864 registered volume values (from the maximum value to a cumulated 1/1000 of
3867 All volumes are in decibels relative to the maximum PCM value.
3869 @subsection Examples
3871 Here is an excerpt of the output:
3873 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3874 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3875 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3876 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3877 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3878 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3879 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3880 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3881 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3887 The mean square energy is approximately -27 dB, or 10^-2.7.
3889 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3891 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3894 In other words, raising the volume by +4 dB does not cause any clipping,
3895 raising it by +5 dB causes clipping for 6 samples, etc.
3897 @c man end AUDIO FILTERS
3899 @chapter Audio Sources
3900 @c man begin AUDIO SOURCES
3902 Below is a description of the currently available audio sources.
3906 Buffer audio frames, and make them available to the filter chain.
3908 This source is mainly intended for a programmatic use, in particular
3909 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3911 It accepts the following parameters:
3915 The timebase which will be used for timestamps of submitted frames. It must be
3916 either a floating-point number or in @var{numerator}/@var{denominator} form.
3919 The sample rate of the incoming audio buffers.
3922 The sample format of the incoming audio buffers.
3923 Either a sample format name or its corresponding integer representation from
3924 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3926 @item channel_layout
3927 The channel layout of the incoming audio buffers.
3928 Either a channel layout name from channel_layout_map in
3929 @file{libavutil/channel_layout.c} or its corresponding integer representation
3930 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3933 The number of channels of the incoming audio buffers.
3934 If both @var{channels} and @var{channel_layout} are specified, then they
3939 @subsection Examples
3942 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3945 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3946 Since the sample format with name "s16p" corresponds to the number
3947 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3950 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3955 Generate an audio signal specified by an expression.
3957 This source accepts in input one or more expressions (one for each
3958 channel), which are evaluated and used to generate a corresponding
3961 This source accepts the following options:
3965 Set the '|'-separated expressions list for each separate channel. In case the
3966 @option{channel_layout} option is not specified, the selected channel layout
3967 depends on the number of provided expressions. Otherwise the last
3968 specified expression is applied to the remaining output channels.
3970 @item channel_layout, c
3971 Set the channel layout. The number of channels in the specified layout
3972 must be equal to the number of specified expressions.
3975 Set the minimum duration of the sourced audio. See
3976 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3977 for the accepted syntax.
3978 Note that the resulting duration may be greater than the specified
3979 duration, as the generated audio is always cut at the end of a
3982 If not specified, or the expressed duration is negative, the audio is
3983 supposed to be generated forever.
3986 Set the number of samples per channel per each output frame,
3989 @item sample_rate, s
3990 Specify the sample rate, default to 44100.
3993 Each expression in @var{exprs} can contain the following constants:
3997 number of the evaluated sample, starting from 0
4000 time of the evaluated sample expressed in seconds, starting from 0
4007 @subsection Examples
4017 Generate a sin signal with frequency of 440 Hz, set sample rate to
4020 aevalsrc="sin(440*2*PI*t):s=8000"
4024 Generate a two channels signal, specify the channel layout (Front
4025 Center + Back Center) explicitly:
4027 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4031 Generate white noise:
4033 aevalsrc="-2+random(0)"
4037 Generate an amplitude modulated signal:
4039 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4043 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4045 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4052 The null audio source, return unprocessed audio frames. It is mainly useful
4053 as a template and to be employed in analysis / debugging tools, or as
4054 the source for filters which ignore the input data (for example the sox
4057 This source accepts the following options:
4061 @item channel_layout, cl
4063 Specifies the channel layout, and can be either an integer or a string
4064 representing a channel layout. The default value of @var{channel_layout}
4067 Check the channel_layout_map definition in
4068 @file{libavutil/channel_layout.c} for the mapping between strings and
4069 channel layout values.
4071 @item sample_rate, r
4072 Specifies the sample rate, and defaults to 44100.
4075 Set the number of samples per requested frames.
4079 @subsection Examples
4083 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4085 anullsrc=r=48000:cl=4
4089 Do the same operation with a more obvious syntax:
4091 anullsrc=r=48000:cl=mono
4095 All the parameters need to be explicitly defined.
4099 Synthesize a voice utterance using the libflite library.
4101 To enable compilation of this filter you need to configure FFmpeg with
4102 @code{--enable-libflite}.
4104 Note that the flite library is not thread-safe.
4106 The filter accepts the following options:
4111 If set to 1, list the names of the available voices and exit
4112 immediately. Default value is 0.
4115 Set the maximum number of samples per frame. Default value is 512.
4118 Set the filename containing the text to speak.
4121 Set the text to speak.
4124 Set the voice to use for the speech synthesis. Default value is
4125 @code{kal}. See also the @var{list_voices} option.
4128 @subsection Examples
4132 Read from file @file{speech.txt}, and synthesize the text using the
4133 standard flite voice:
4135 flite=textfile=speech.txt
4139 Read the specified text selecting the @code{slt} voice:
4141 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4145 Input text to ffmpeg:
4147 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4151 Make @file{ffplay} speak the specified text, using @code{flite} and
4152 the @code{lavfi} device:
4154 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4158 For more information about libflite, check:
4159 @url{http://www.speech.cs.cmu.edu/flite/}
4163 Generate a noise audio signal.
4165 The filter accepts the following options:
4168 @item sample_rate, r
4169 Specify the sample rate. Default value is 48000 Hz.
4172 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4176 Specify the duration of the generated audio stream. Not specifying this option
4177 results in noise with an infinite length.
4179 @item color, colour, c
4180 Specify the color of noise. Available noise colors are white, pink, and brown.
4181 Default color is white.
4184 Specify a value used to seed the PRNG.
4187 Set the number of samples per each output frame, default is 1024.
4190 @subsection Examples
4195 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4197 anoisesrc=d=60:c=pink:r=44100:a=0.5
4203 Generate an audio signal made of a sine wave with amplitude 1/8.
4205 The audio signal is bit-exact.
4207 The filter accepts the following options:
4212 Set the carrier frequency. Default is 440 Hz.
4214 @item beep_factor, b
4215 Enable a periodic beep every second with frequency @var{beep_factor} times
4216 the carrier frequency. Default is 0, meaning the beep is disabled.
4218 @item sample_rate, r
4219 Specify the sample rate, default is 44100.
4222 Specify the duration of the generated audio stream.
4224 @item samples_per_frame
4225 Set the number of samples per output frame.
4227 The expression can contain the following constants:
4231 The (sequential) number of the output audio frame, starting from 0.
4234 The PTS (Presentation TimeStamp) of the output audio frame,
4235 expressed in @var{TB} units.
4238 The PTS of the output audio frame, expressed in seconds.
4241 The timebase of the output audio frames.
4244 Default is @code{1024}.
4247 @subsection Examples
4252 Generate a simple 440 Hz sine wave:
4258 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4262 sine=frequency=220:beep_factor=4:duration=5
4266 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4269 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4273 @c man end AUDIO SOURCES
4275 @chapter Audio Sinks
4276 @c man begin AUDIO SINKS
4278 Below is a description of the currently available audio sinks.
4280 @section abuffersink
4282 Buffer audio frames, and make them available to the end of filter chain.
4284 This sink is mainly intended for programmatic use, in particular
4285 through the interface defined in @file{libavfilter/buffersink.h}
4286 or the options system.
4288 It accepts a pointer to an AVABufferSinkContext structure, which
4289 defines the incoming buffers' formats, to be passed as the opaque
4290 parameter to @code{avfilter_init_filter} for initialization.
4293 Null audio sink; do absolutely nothing with the input audio. It is
4294 mainly useful as a template and for use in analysis / debugging
4297 @c man end AUDIO SINKS
4299 @chapter Video Filters
4300 @c man begin VIDEO FILTERS
4302 When you configure your FFmpeg build, you can disable any of the
4303 existing filters using @code{--disable-filters}.
4304 The configure output will show the video filters included in your
4307 Below is a description of the currently available video filters.
4309 @section alphaextract
4311 Extract the alpha component from the input as a grayscale video. This
4312 is especially useful with the @var{alphamerge} filter.
4316 Add or replace the alpha component of the primary input with the
4317 grayscale value of a second input. This is intended for use with
4318 @var{alphaextract} to allow the transmission or storage of frame
4319 sequences that have alpha in a format that doesn't support an alpha
4322 For example, to reconstruct full frames from a normal YUV-encoded video
4323 and a separate video created with @var{alphaextract}, you might use:
4325 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4328 Since this filter is designed for reconstruction, it operates on frame
4329 sequences without considering timestamps, and terminates when either
4330 input reaches end of stream. This will cause problems if your encoding
4331 pipeline drops frames. If you're trying to apply an image as an
4332 overlay to a video stream, consider the @var{overlay} filter instead.
4336 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4337 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4338 Substation Alpha) subtitles files.
4340 This filter accepts the following option in addition to the common options from
4341 the @ref{subtitles} filter:
4345 Set the shaping engine
4347 Available values are:
4350 The default libass shaping engine, which is the best available.
4352 Fast, font-agnostic shaper that can do only substitutions
4354 Slower shaper using OpenType for substitutions and positioning
4357 The default is @code{auto}.
4361 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4363 The filter accepts the following options:
4367 Set threshold A for 1st plane. Default is 0.02.
4368 Valid range is 0 to 0.3.
4371 Set threshold B for 1st plane. Default is 0.04.
4372 Valid range is 0 to 5.
4375 Set threshold A for 2nd plane. Default is 0.02.
4376 Valid range is 0 to 0.3.
4379 Set threshold B for 2nd plane. Default is 0.04.
4380 Valid range is 0 to 5.
4383 Set threshold A for 3rd plane. Default is 0.02.
4384 Valid range is 0 to 0.3.
4387 Set threshold B for 3rd plane. Default is 0.04.
4388 Valid range is 0 to 5.
4390 Threshold A is designed to react on abrupt changes in the input signal and
4391 threshold B is designed to react on continuous changes in the input signal.
4394 Set number of frames filter will use for averaging. Default is 33. Must be odd
4395 number in range [5, 129].
4398 Set what planes of frame filter will use for averaging. Default is all.
4403 Apply average blur filter.
4405 The filter accepts the following options:
4409 Set horizontal kernel size.
4412 Set which planes to filter. By default all planes are filtered.
4415 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4416 Default is @code{0}.
4421 Compute the bounding box for the non-black pixels in the input frame
4424 This filter computes the bounding box containing all the pixels with a
4425 luminance value greater than the minimum allowed value.
4426 The parameters describing the bounding box are printed on the filter
4429 The filter accepts the following option:
4433 Set the minimal luminance value. Default is @code{16}.
4436 @section bitplanenoise
4438 Show and measure bit plane noise.
4440 The filter accepts the following options:
4444 Set which plane to analyze. Default is @code{1}.
4447 Filter out noisy pixels from @code{bitplane} set above.
4448 Default is disabled.
4451 @section blackdetect
4453 Detect video intervals that are (almost) completely black. Can be
4454 useful to detect chapter transitions, commercials, or invalid
4455 recordings. Output lines contains the time for the start, end and
4456 duration of the detected black interval expressed in seconds.
4458 In order to display the output lines, you need to set the loglevel at
4459 least to the AV_LOG_INFO value.
4461 The filter accepts the following options:
4464 @item black_min_duration, d
4465 Set the minimum detected black duration expressed in seconds. It must
4466 be a non-negative floating point number.
4468 Default value is 2.0.
4470 @item picture_black_ratio_th, pic_th
4471 Set the threshold for considering a picture "black".
4472 Express the minimum value for the ratio:
4474 @var{nb_black_pixels} / @var{nb_pixels}
4477 for which a picture is considered black.
4478 Default value is 0.98.
4480 @item pixel_black_th, pix_th
4481 Set the threshold for considering a pixel "black".
4483 The threshold expresses the maximum pixel luminance value for which a
4484 pixel is considered "black". The provided value is scaled according to
4485 the following equation:
4487 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4490 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4491 the input video format, the range is [0-255] for YUV full-range
4492 formats and [16-235] for YUV non full-range formats.
4494 Default value is 0.10.
4497 The following example sets the maximum pixel threshold to the minimum
4498 value, and detects only black intervals of 2 or more seconds:
4500 blackdetect=d=2:pix_th=0.00
4505 Detect frames that are (almost) completely black. Can be useful to
4506 detect chapter transitions or commercials. Output lines consist of
4507 the frame number of the detected frame, the percentage of blackness,
4508 the position in the file if known or -1 and the timestamp in seconds.
4510 In order to display the output lines, you need to set the loglevel at
4511 least to the AV_LOG_INFO value.
4513 This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4514 The value represents the percentage of pixels in the picture that
4515 are below the threshold value.
4517 It accepts the following parameters:
4522 The percentage of the pixels that have to be below the threshold; it defaults to
4525 @item threshold, thresh
4526 The threshold below which a pixel value is considered black; it defaults to
4531 @section blend, tblend
4533 Blend two video frames into each other.
4535 The @code{blend} filter takes two input streams and outputs one
4536 stream, the first input is the "top" layer and second input is
4537 "bottom" layer. By default, the output terminates when the longest input terminates.
4539 The @code{tblend} (time blend) filter takes two consecutive frames
4540 from one single stream, and outputs the result obtained by blending
4541 the new frame on top of the old frame.
4543 A description of the accepted options follows.
4551 Set blend mode for specific pixel component or all pixel components in case
4552 of @var{all_mode}. Default value is @code{normal}.
4554 Available values for component modes are:
4595 Set blend opacity for specific pixel component or all pixel components in case
4596 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4603 Set blend expression for specific pixel component or all pixel components in case
4604 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4606 The expressions can use the following variables:
4610 The sequential number of the filtered frame, starting from @code{0}.
4614 the coordinates of the current sample
4618 the width and height of currently filtered plane
4622 Width and height scale depending on the currently filtered plane. It is the
4623 ratio between the corresponding luma plane number of pixels and the current
4624 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4625 @code{0.5,0.5} for chroma planes.
4628 Time of the current frame, expressed in seconds.
4631 Value of pixel component at current location for first video frame (top layer).
4634 Value of pixel component at current location for second video frame (bottom layer).
4638 Force termination when the shortest input terminates. Default is
4639 @code{0}. This option is only defined for the @code{blend} filter.
4642 Continue applying the last bottom frame after the end of the stream. A value of
4643 @code{0} disable the filter after the last frame of the bottom layer is reached.
4644 Default is @code{1}. This option is only defined for the @code{blend} filter.
4647 @subsection Examples
4651 Apply transition from bottom layer to top layer in first 10 seconds:
4653 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4657 Apply 1x1 checkerboard effect:
4659 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4663 Apply uncover left effect:
4665 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4669 Apply uncover down effect:
4671 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4675 Apply uncover up-left effect:
4677 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4681 Split diagonally video and shows top and bottom layer on each side:
4683 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4687 Display differences between the current and the previous frame:
4689 tblend=all_mode=difference128
4695 Apply a boxblur algorithm to the input video.
4697 It accepts the following parameters:
4701 @item luma_radius, lr
4702 @item luma_power, lp
4703 @item chroma_radius, cr
4704 @item chroma_power, cp
4705 @item alpha_radius, ar
4706 @item alpha_power, ap
4710 A description of the accepted options follows.
4713 @item luma_radius, lr
4714 @item chroma_radius, cr
4715 @item alpha_radius, ar
4716 Set an expression for the box radius in pixels used for blurring the
4717 corresponding input plane.
4719 The radius value must be a non-negative number, and must not be
4720 greater than the value of the expression @code{min(w,h)/2} for the
4721 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4724 Default value for @option{luma_radius} is "2". If not specified,
4725 @option{chroma_radius} and @option{alpha_radius} default to the
4726 corresponding value set for @option{luma_radius}.
4728 The expressions can contain the following constants:
4732 The input width and height in pixels.
4736 The input chroma image width and height in pixels.
4740 The horizontal and vertical chroma subsample values. For example, for the
4741 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4744 @item luma_power, lp
4745 @item chroma_power, cp
4746 @item alpha_power, ap
4747 Specify how many times the boxblur filter is applied to the
4748 corresponding plane.
4750 Default value for @option{luma_power} is 2. If not specified,
4751 @option{chroma_power} and @option{alpha_power} default to the
4752 corresponding value set for @option{luma_power}.
4754 A value of 0 will disable the effect.
4757 @subsection Examples
4761 Apply a boxblur filter with the luma, chroma, and alpha radii
4764 boxblur=luma_radius=2:luma_power=1
4769 Set the luma radius to 2, and alpha and chroma radius to 0:
4771 boxblur=2:1:cr=0:ar=0
4775 Set the luma and chroma radii to a fraction of the video dimension:
4777 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4783 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4784 Deinterlacing Filter").
4786 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4787 interpolation algorithms.
4788 It accepts the following parameters:
4792 The interlacing mode to adopt. It accepts one of the following values:
4796 Output one frame for each frame.
4798 Output one frame for each field.
4801 The default value is @code{send_field}.
4804 The picture field parity assumed for the input interlaced video. It accepts one
4805 of the following values:
4809 Assume the top field is first.
4811 Assume the bottom field is first.
4813 Enable automatic detection of field parity.
4816 The default value is @code{auto}.
4817 If the interlacing is unknown or the decoder does not export this information,
4818 top field first will be assumed.
4821 Specify which frames to deinterlace. Accept one of the following
4826 Deinterlace all frames.
4828 Only deinterlace frames marked as interlaced.
4831 The default value is @code{all}.
4835 YUV colorspace color/chroma keying.
4837 The filter accepts the following options:
4841 The color which will be replaced with transparency.
4844 Similarity percentage with the key color.
4846 0.01 matches only the exact key color, while 1.0 matches everything.
4851 0.0 makes pixels either fully transparent, or not transparent at all.
4853 Higher values result in semi-transparent pixels, with a higher transparency
4854 the more similar the pixels color is to the key color.
4857 Signals that the color passed is already in YUV instead of RGB.
4859 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4860 This can be used to pass exact YUV values as hexadecimal numbers.
4863 @subsection Examples
4867 Make every green pixel in the input image transparent:
4869 ffmpeg -i input.png -vf chromakey=green out.png
4873 Overlay a greenscreen-video on top of a static black background.
4875 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
4881 Display CIE color diagram with pixels overlaid onto it.
4883 The filter accepts the following options:
4898 @item uhdtv, rec2020
4911 Set what gamuts to draw.
4913 See @code{system} option for available values.
4916 Set ciescope size, by default set to 512.
4919 Set intensity used to map input pixel values to CIE diagram.
4922 Set contrast used to draw tongue colors that are out of active color system gamut.
4925 Correct gamma displayed on scope, by default enabled.
4928 Show white point on CIE diagram, by default disabled.
4931 Set input gamma. Used only with XYZ input color space.
4936 Visualize information exported by some codecs.
4938 Some codecs can export information through frames using side-data or other
4939 means. For example, some MPEG based codecs export motion vectors through the
4940 @var{export_mvs} flag in the codec @option{flags2} option.
4942 The filter accepts the following option:
4946 Set motion vectors to visualize.
4948 Available flags for @var{mv} are:
4952 forward predicted MVs of P-frames
4954 forward predicted MVs of B-frames
4956 backward predicted MVs of B-frames
4960 Display quantization parameters using the chroma planes.
4963 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4965 Available flags for @var{mv_type} are:
4969 forward predicted MVs
4971 backward predicted MVs
4974 @item frame_type, ft
4975 Set frame type to visualize motion vectors of.
4977 Available flags for @var{frame_type} are:
4981 intra-coded frames (I-frames)
4983 predicted frames (P-frames)
4985 bi-directionally predicted frames (B-frames)
4989 @subsection Examples
4993 Visualize forward predicted MVs of all frames using @command{ffplay}:
4995 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
4999 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5001 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5005 @section colorbalance
5006 Modify intensity of primary colors (red, green and blue) of input frames.
5008 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5009 regions for the red-cyan, green-magenta or blue-yellow balance.
5011 A positive adjustment value shifts the balance towards the primary color, a negative
5012 value towards the complementary color.
5014 The filter accepts the following options:
5020 Adjust red, green and blue shadows (darkest pixels).
5025 Adjust red, green and blue midtones (medium pixels).
5030 Adjust red, green and blue highlights (brightest pixels).
5032 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5035 @subsection Examples
5039 Add red color cast to shadows:
5046 RGB colorspace color keying.
5048 The filter accepts the following options:
5052 The color which will be replaced with transparency.
5055 Similarity percentage with the key color.
5057 0.01 matches only the exact key color, while 1.0 matches everything.
5062 0.0 makes pixels either fully transparent, or not transparent at all.
5064 Higher values result in semi-transparent pixels, with a higher transparency
5065 the more similar the pixels color is to the key color.
5068 @subsection Examples
5072 Make every green pixel in the input image transparent:
5074 ffmpeg -i input.png -vf colorkey=green out.png
5078 Overlay a greenscreen-video on top of a static background image.
5080 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
5084 @section colorlevels
5086 Adjust video input frames using levels.
5088 The filter accepts the following options:
5095 Adjust red, green, blue and alpha input black point.
5096 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5102 Adjust red, green, blue and alpha input white point.
5103 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
5105 Input levels are used to lighten highlights (bright tones), darken shadows
5106 (dark tones), change the balance of bright and dark tones.
5112 Adjust red, green, blue and alpha output black point.
5113 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5119 Adjust red, green, blue and alpha output white point.
5120 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5122 Output levels allows manual selection of a constrained output level range.
5125 @subsection Examples
5129 Make video output darker:
5131 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5137 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5141 Make video output lighter:
5143 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5147 Increase brightness:
5149 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5153 @section colorchannelmixer
5155 Adjust video input frames by re-mixing color channels.
5157 This filter modifies a color channel by adding the values associated to
5158 the other channels of the same pixels. For example if the value to
5159 modify is red, the output value will be:
5161 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5164 The filter accepts the following options:
5171 Adjust contribution of input red, green, blue and alpha channels for output red channel.
5172 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5178 Adjust contribution of input red, green, blue and alpha channels for output green channel.
5179 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5185 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5186 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5192 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5193 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5195 Allowed ranges for options are @code{[-2.0, 2.0]}.
5198 @subsection Examples
5202 Convert source to grayscale:
5204 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5207 Simulate sepia tones:
5209 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5213 @section colormatrix
5215 Convert color matrix.
5217 The filter accepts the following options:
5222 Specify the source and destination color matrix. Both values must be
5225 The accepted values are:
5253 For example to convert from BT.601 to SMPTE-240M, use the command:
5255 colormatrix=bt601:smpte240m
5260 Convert colorspace, transfer characteristics or color primaries.
5261 Input video needs to have an even size.
5263 The filter accepts the following options:
5268 Specify all color properties at once.
5270 The accepted values are:
5300 Specify output colorspace.
5302 The accepted values are:
5311 BT.470BG or BT.601-6 625
5314 SMPTE-170M or BT.601-6 525
5323 BT.2020 with non-constant luminance
5329 Specify output transfer characteristics.
5331 The accepted values are:
5343 Constant gamma of 2.2
5346 Constant gamma of 2.8
5349 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5367 BT.2020 for 10-bits content
5370 BT.2020 for 12-bits content
5376 Specify output color primaries.
5378 The accepted values are:
5387 BT.470BG or BT.601-6 625
5390 SMPTE-170M or BT.601-6 525
5411 Specify output color range.
5413 The accepted values are:
5416 TV (restricted) range
5419 MPEG (restricted) range
5430 Specify output color format.
5432 The accepted values are:
5435 YUV 4:2:0 planar 8-bits
5438 YUV 4:2:0 planar 10-bits
5441 YUV 4:2:0 planar 12-bits
5444 YUV 4:2:2 planar 8-bits
5447 YUV 4:2:2 planar 10-bits
5450 YUV 4:2:2 planar 12-bits
5453 YUV 4:4:4 planar 8-bits
5456 YUV 4:4:4 planar 10-bits
5459 YUV 4:4:4 planar 12-bits
5464 Do a fast conversion, which skips gamma/primary correction. This will take
5465 significantly less CPU, but will be mathematically incorrect. To get output
5466 compatible with that produced by the colormatrix filter, use fast=1.
5469 Specify dithering mode.
5471 The accepted values are:
5477 Floyd-Steinberg dithering
5481 Whitepoint adaptation mode.
5483 The accepted values are:
5486 Bradford whitepoint adaptation
5489 von Kries whitepoint adaptation
5492 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5496 Override all input properties at once. Same accepted values as @ref{all}.
5499 Override input colorspace. Same accepted values as @ref{space}.
5502 Override input color primaries. Same accepted values as @ref{primaries}.
5505 Override input transfer characteristics. Same accepted values as @ref{trc}.
5508 Override input color range. Same accepted values as @ref{range}.
5512 The filter converts the transfer characteristics, color space and color
5513 primaries to the specified user values. The output value, if not specified,
5514 is set to a default value based on the "all" property. If that property is
5515 also not specified, the filter will log an error. The output color range and
5516 format default to the same value as the input color range and format. The
5517 input transfer characteristics, color space, color primaries and color range
5518 should be set on the input data. If any of these are missing, the filter will
5519 log an error and no conversion will take place.
5521 For example to convert the input to SMPTE-240M, use the command:
5523 colorspace=smpte240m
5526 @section convolution
5528 Apply convolution 3x3 or 5x5 filter.
5530 The filter accepts the following options:
5537 Set matrix for each plane.
5538 Matrix is sequence of 9 or 25 signed integers.
5544 Set multiplier for calculated value for each plane.
5550 Set bias for each plane. This value is added to the result of the multiplication.
5551 Useful for making the overall image brighter or darker. Default is 0.0.
5554 @subsection Examples
5560 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"
5566 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"
5572 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"
5578 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"
5584 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"
5590 Copy the input source unchanged to the output. This is mainly useful for
5595 Video filtering on GPU using Apple's CoreImage API on OSX.
5597 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5598 processed by video hardware. However, software-based OpenGL implementations
5599 exist which means there is no guarantee for hardware processing. It depends on
5602 There are many filters and image generators provided by Apple that come with a
5603 large variety of options. The filter has to be referenced by its name along
5606 The coreimage filter accepts the following options:
5609 List all available filters and generators along with all their respective
5610 options as well as possible minimum and maximum values along with the default
5617 Specify all filters by their respective name and options.
5618 Use @var{list_filters} to determine all valid filter names and options.
5619 Numerical options are specified by a float value and are automatically clamped
5620 to their respective value range. Vector and color options have to be specified
5621 by a list of space separated float values. Character escaping has to be done.
5622 A special option name @code{default} is available to use default options for a
5625 It is required to specify either @code{default} or at least one of the filter options.
5626 All omitted options are used with their default values.
5627 The syntax of the filter string is as follows:
5629 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5633 Specify a rectangle where the output of the filter chain is copied into the
5634 input image. It is given by a list of space separated float values:
5636 output_rect=x\ y\ width\ height
5638 If not given, the output rectangle equals the dimensions of the input image.
5639 The output rectangle is automatically cropped at the borders of the input
5640 image. Negative values are valid for each component.
5642 output_rect=25\ 25\ 100\ 100
5646 Several filters can be chained for successive processing without GPU-HOST
5647 transfers allowing for fast processing of complex filter chains.
5648 Currently, only filters with zero (generators) or exactly one (filters) input
5649 image and one output image are supported. Also, transition filters are not yet
5652 Some filters generate output images with additional padding depending on the
5653 respective filter kernel. The padding is automatically removed to ensure the
5654 filter output has the same size as the input image.
5656 For image generators, the size of the output image is determined by the
5657 previous output image of the filter chain or the input image of the whole
5658 filterchain, respectively. The generators do not use the pixel information of
5659 this image to generate their output. However, the generated output is
5660 blended onto this image, resulting in partial or complete coverage of the
5663 The @ref{coreimagesrc} video source can be used for generating input images
5664 which are directly fed into the filter chain. By using it, providing input
5665 images by another video source or an input video is not required.
5667 @subsection Examples
5672 List all filters available:
5674 coreimage=list_filters=true
5678 Use the CIBoxBlur filter with default options to blur an image:
5680 coreimage=filter=CIBoxBlur@@default
5684 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5685 its center at 100x100 and a radius of 50 pixels:
5687 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5691 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5692 given as complete and escaped command-line for Apple's standard bash shell:
5694 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5700 Crop the input video to given dimensions.
5702 It accepts the following parameters:
5706 The width of the output video. It defaults to @code{iw}.
5707 This expression is evaluated only once during the filter
5708 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5711 The height of the output video. It defaults to @code{ih}.
5712 This expression is evaluated only once during the filter
5713 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5716 The horizontal position, in the input video, of the left edge of the output
5717 video. It defaults to @code{(in_w-out_w)/2}.
5718 This expression is evaluated per-frame.
5721 The vertical position, in the input video, of the top edge of the output video.
5722 It defaults to @code{(in_h-out_h)/2}.
5723 This expression is evaluated per-frame.
5726 If set to 1 will force the output display aspect ratio
5727 to be the same of the input, by changing the output sample aspect
5728 ratio. It defaults to 0.
5731 Enable exact cropping. If enabled, subsampled videos will be cropped at exact
5732 width/height/x/y as specified and will not be rounded to nearest smaller value.
5736 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5737 expressions containing the following constants:
5742 The computed values for @var{x} and @var{y}. They are evaluated for
5747 The input width and height.
5751 These are the same as @var{in_w} and @var{in_h}.
5755 The output (cropped) width and height.
5759 These are the same as @var{out_w} and @var{out_h}.
5762 same as @var{iw} / @var{ih}
5765 input sample aspect ratio
5768 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5772 horizontal and vertical chroma subsample values. For example for the
5773 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5776 The number of the input frame, starting from 0.
5779 the position in the file of the input frame, NAN if unknown
5782 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5786 The expression for @var{out_w} may depend on the value of @var{out_h},
5787 and the expression for @var{out_h} may depend on @var{out_w}, but they
5788 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5789 evaluated after @var{out_w} and @var{out_h}.
5791 The @var{x} and @var{y} parameters specify the expressions for the
5792 position of the top-left corner of the output (non-cropped) area. They
5793 are evaluated for each frame. If the evaluated value is not valid, it
5794 is approximated to the nearest valid value.
5796 The expression for @var{x} may depend on @var{y}, and the expression
5797 for @var{y} may depend on @var{x}.
5799 @subsection Examples
5803 Crop area with size 100x100 at position (12,34).
5808 Using named options, the example above becomes:
5810 crop=w=100:h=100:x=12:y=34
5814 Crop the central input area with size 100x100:
5820 Crop the central input area with size 2/3 of the input video:
5822 crop=2/3*in_w:2/3*in_h
5826 Crop the input video central square:
5833 Delimit the rectangle with the top-left corner placed at position
5834 100:100 and the right-bottom corner corresponding to the right-bottom
5835 corner of the input image.
5837 crop=in_w-100:in_h-100:100:100
5841 Crop 10 pixels from the left and right borders, and 20 pixels from
5842 the top and bottom borders
5844 crop=in_w-2*10:in_h-2*20
5848 Keep only the bottom right quarter of the input image:
5850 crop=in_w/2:in_h/2:in_w/2:in_h/2
5854 Crop height for getting Greek harmony:
5856 crop=in_w:1/PHI*in_w
5860 Apply trembling effect:
5862 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)
5866 Apply erratic camera effect depending on timestamp:
5868 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)"
5872 Set x depending on the value of y:
5874 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5878 @subsection Commands
5880 This filter supports the following commands:
5886 Set width/height of the output video and the horizontal/vertical position
5888 The command accepts the same syntax of the corresponding option.
5890 If the specified expression is not valid, it is kept at its current
5896 Auto-detect the crop size.
5898 It calculates the necessary cropping parameters and prints the
5899 recommended parameters via the logging system. The detected dimensions
5900 correspond to the non-black area of the input video.
5902 It accepts the following parameters:
5907 Set higher black value threshold, which can be optionally specified
5908 from nothing (0) to everything (255 for 8-bit based formats). An intensity
5909 value greater to the set value is considered non-black. It defaults to 24.
5910 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5911 on the bitdepth of the pixel format.
5914 The value which the width/height should be divisible by. It defaults to
5915 16. The offset is automatically adjusted to center the video. Use 2 to
5916 get only even dimensions (needed for 4:2:2 video). 16 is best when
5917 encoding to most video codecs.
5919 @item reset_count, reset
5920 Set the counter that determines after how many frames cropdetect will
5921 reset the previously detected largest video area and start over to
5922 detect the current optimal crop area. Default value is 0.
5924 This can be useful when channel logos distort the video area. 0
5925 indicates 'never reset', and returns the largest area encountered during
5932 Apply color adjustments using curves.
5934 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5935 component (red, green and blue) has its values defined by @var{N} key points
5936 tied from each other using a smooth curve. The x-axis represents the pixel
5937 values from the input frame, and the y-axis the new pixel values to be set for
5940 By default, a component curve is defined by the two points @var{(0;0)} and
5941 @var{(1;1)}. This creates a straight line where each original pixel value is
5942 "adjusted" to its own value, which means no change to the image.
5944 The filter allows you to redefine these two points and add some more. A new
5945 curve (using a natural cubic spline interpolation) will be define to pass
5946 smoothly through all these new coordinates. The new defined points needs to be
5947 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5948 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5949 the vector spaces, the values will be clipped accordingly.
5951 The filter accepts the following options:
5955 Select one of the available color presets. This option can be used in addition
5956 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5957 options takes priority on the preset values.
5958 Available presets are:
5961 @item color_negative
5964 @item increase_contrast
5966 @item linear_contrast
5967 @item medium_contrast
5969 @item strong_contrast
5972 Default is @code{none}.
5974 Set the master key points. These points will define a second pass mapping. It
5975 is sometimes called a "luminance" or "value" mapping. It can be used with
5976 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
5977 post-processing LUT.
5979 Set the key points for the red component.
5981 Set the key points for the green component.
5983 Set the key points for the blue component.
5985 Set the key points for all components (not including master).
5986 Can be used in addition to the other key points component
5987 options. In this case, the unset component(s) will fallback on this
5988 @option{all} setting.
5990 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
5992 Save Gnuplot script of the curves in specified file.
5995 To avoid some filtergraph syntax conflicts, each key points list need to be
5996 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
5998 @subsection Examples
6002 Increase slightly the middle level of blue:
6004 curves=blue='0/0 0.5/0.58 1/1'
6010 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'
6012 Here we obtain the following coordinates for each components:
6015 @code{(0;0.11) (0.42;0.51) (1;0.95)}
6017 @code{(0;0) (0.50;0.48) (1;1)}
6019 @code{(0;0.22) (0.49;0.44) (1;0.80)}
6023 The previous example can also be achieved with the associated built-in preset:
6025 curves=preset=vintage
6035 Use a Photoshop preset and redefine the points of the green component:
6037 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6041 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6042 and @command{gnuplot}:
6044 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
6045 gnuplot -p /tmp/curves.plt
6051 Video data analysis filter.
6053 This filter shows hexadecimal pixel values of part of video.
6055 The filter accepts the following options:
6059 Set output video size.
6062 Set x offset from where to pick pixels.
6065 Set y offset from where to pick pixels.
6068 Set scope mode, can be one of the following:
6071 Draw hexadecimal pixel values with white color on black background.
6074 Draw hexadecimal pixel values with input video pixel color on black
6078 Draw hexadecimal pixel values on color background picked from input video,
6079 the text color is picked in such way so its always visible.
6083 Draw rows and columns numbers on left and top of video.
6086 Set background opacity.
6091 Denoise frames using 2D DCT (frequency domain filtering).
6093 This filter is not designed for real time.
6095 The filter accepts the following options:
6099 Set the noise sigma constant.
6101 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6102 coefficient (absolute value) below this threshold with be dropped.
6104 If you need a more advanced filtering, see @option{expr}.
6106 Default is @code{0}.
6109 Set number overlapping pixels for each block. Since the filter can be slow, you
6110 may want to reduce this value, at the cost of a less effective filter and the
6111 risk of various artefacts.
6113 If the overlapping value doesn't permit processing the whole input width or
6114 height, a warning will be displayed and according borders won't be denoised.
6116 Default value is @var{blocksize}-1, which is the best possible setting.
6119 Set the coefficient factor expression.
6121 For each coefficient of a DCT block, this expression will be evaluated as a
6122 multiplier value for the coefficient.
6124 If this is option is set, the @option{sigma} option will be ignored.
6126 The absolute value of the coefficient can be accessed through the @var{c}
6130 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
6131 @var{blocksize}, which is the width and height of the processed blocks.
6133 The default value is @var{3} (8x8) and can be raised to @var{4} for a
6134 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
6135 on the speed processing. Also, a larger block size does not necessarily means a
6139 @subsection Examples
6141 Apply a denoise with a @option{sigma} of @code{4.5}:
6146 The same operation can be achieved using the expression system:
6148 dctdnoiz=e='gte(c, 4.5*3)'
6151 Violent denoise using a block size of @code{16x16}:
6158 Remove banding artifacts from input video.
6159 It works by replacing banded pixels with average value of referenced pixels.
6161 The filter accepts the following options:
6168 Set banding detection threshold for each plane. Default is 0.02.
6169 Valid range is 0.00003 to 0.5.
6170 If difference between current pixel and reference pixel is less than threshold,
6171 it will be considered as banded.
6174 Banding detection range in pixels. Default is 16. If positive, random number
6175 in range 0 to set value will be used. If negative, exact absolute value
6177 The range defines square of four pixels around current pixel.
6180 Set direction in radians from which four pixel will be compared. If positive,
6181 random direction from 0 to set direction will be picked. If negative, exact of
6182 absolute value will be picked. For example direction 0, -PI or -2*PI radians
6183 will pick only pixels on same row and -PI/2 will pick only pixels on same
6187 If enabled, current pixel is compared with average value of all four
6188 surrounding pixels. The default is enabled. If disabled current pixel is
6189 compared with all four surrounding pixels. The pixel is considered banded
6190 if only all four differences with surrounding pixels are less than threshold.
6193 If enabled, current pixel is changed if and only if all pixel components are banded,
6194 e.g. banding detection threshold is triggered for all color components.
6195 The default is disabled.
6201 Drop duplicated frames at regular intervals.
6203 The filter accepts the following options:
6207 Set the number of frames from which one will be dropped. Setting this to
6208 @var{N} means one frame in every batch of @var{N} frames will be dropped.
6209 Default is @code{5}.
6212 Set the threshold for duplicate detection. If the difference metric for a frame
6213 is less than or equal to this value, then it is declared as duplicate. Default
6217 Set scene change threshold. Default is @code{15}.
6221 Set the size of the x and y-axis blocks used during metric calculations.
6222 Larger blocks give better noise suppression, but also give worse detection of
6223 small movements. Must be a power of two. Default is @code{32}.
6226 Mark main input as a pre-processed input and activate clean source input
6227 stream. This allows the input to be pre-processed with various filters to help
6228 the metrics calculation while keeping the frame selection lossless. When set to
6229 @code{1}, the first stream is for the pre-processed input, and the second
6230 stream is the clean source from where the kept frames are chosen. Default is
6234 Set whether or not chroma is considered in the metric calculations. Default is
6240 Apply deflate effect to the video.
6242 This filter replaces the pixel by the local(3x3) average by taking into account
6243 only values lower than the pixel.
6245 It accepts the following options:
6252 Limit the maximum change for each plane, default is 65535.
6253 If 0, plane will remain unchanged.
6258 Remove judder produced by partially interlaced telecined content.
6260 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6261 source was partially telecined content then the output of @code{pullup,dejudder}
6262 will have a variable frame rate. May change the recorded frame rate of the
6263 container. Aside from that change, this filter will not affect constant frame
6266 The option available in this filter is:
6270 Specify the length of the window over which the judder repeats.
6272 Accepts any integer greater than 1. Useful values are:
6276 If the original was telecined from 24 to 30 fps (Film to NTSC).
6279 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6282 If a mixture of the two.
6285 The default is @samp{4}.
6290 Suppress a TV station logo by a simple interpolation of the surrounding
6291 pixels. Just set a rectangle covering the logo and watch it disappear
6292 (and sometimes something even uglier appear - your mileage may vary).
6294 It accepts the following parameters:
6299 Specify the top left corner coordinates of the logo. They must be
6304 Specify the width and height of the logo to clear. They must be
6308 Specify the thickness of the fuzzy edge of the rectangle (added to
6309 @var{w} and @var{h}). The default value is 1. This option is
6310 deprecated, setting higher values should no longer be necessary and
6314 When set to 1, a green rectangle is drawn on the screen to simplify
6315 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6316 The default value is 0.
6318 The rectangle is drawn on the outermost pixels which will be (partly)
6319 replaced with interpolated values. The values of the next pixels
6320 immediately outside this rectangle in each direction will be used to
6321 compute the interpolated pixel values inside the rectangle.
6325 @subsection Examples
6329 Set a rectangle covering the area with top left corner coordinates 0,0
6330 and size 100x77, and a band of size 10:
6332 delogo=x=0:y=0:w=100:h=77:band=10
6339 Attempt to fix small changes in horizontal and/or vertical shift. This
6340 filter helps remove camera shake from hand-holding a camera, bumping a
6341 tripod, moving on a vehicle, etc.
6343 The filter accepts the following options:
6351 Specify a rectangular area where to limit the search for motion
6353 If desired the search for motion vectors can be limited to a
6354 rectangular area of the frame defined by its top left corner, width
6355 and height. These parameters have the same meaning as the drawbox
6356 filter which can be used to visualise the position of the bounding
6359 This is useful when simultaneous movement of subjects within the frame
6360 might be confused for camera motion by the motion vector search.
6362 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6363 then the full frame is used. This allows later options to be set
6364 without specifying the bounding box for the motion vector search.
6366 Default - search the whole frame.
6370 Specify the maximum extent of movement in x and y directions in the
6371 range 0-64 pixels. Default 16.
6374 Specify how to generate pixels to fill blanks at the edge of the
6375 frame. Available values are:
6378 Fill zeroes at blank locations
6380 Original image at blank locations
6382 Extruded edge value at blank locations
6384 Mirrored edge at blank locations
6386 Default value is @samp{mirror}.
6389 Specify the blocksize to use for motion search. Range 4-128 pixels,
6393 Specify the contrast threshold for blocks. Only blocks with more than
6394 the specified contrast (difference between darkest and lightest
6395 pixels) will be considered. Range 1-255, default 125.
6398 Specify the search strategy. Available values are:
6401 Set exhaustive search
6403 Set less exhaustive search.
6405 Default value is @samp{exhaustive}.
6408 If set then a detailed log of the motion search is written to the
6412 If set to 1, specify using OpenCL capabilities, only available if
6413 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6419 Apply an exact inverse of the telecine operation. It requires a predefined
6420 pattern specified using the pattern option which must be the same as that passed
6421 to the telecine filter.
6423 This filter accepts the following options:
6432 The default value is @code{top}.
6436 A string of numbers representing the pulldown pattern you wish to apply.
6437 The default value is @code{23}.
6440 A number representing position of the first frame with respect to the telecine
6441 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6446 Apply dilation effect to the video.
6448 This filter replaces the pixel by the local(3x3) maximum.
6450 It accepts the following options:
6457 Limit the maximum change for each plane, default is 65535.
6458 If 0, plane will remain unchanged.
6461 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6464 Flags to local 3x3 coordinates maps like this:
6473 Displace pixels as indicated by second and third input stream.
6475 It takes three input streams and outputs one stream, the first input is the
6476 source, and second and third input are displacement maps.
6478 The second input specifies how much to displace pixels along the
6479 x-axis, while the third input specifies how much to displace pixels
6481 If one of displacement map streams terminates, last frame from that
6482 displacement map will be used.
6484 Note that once generated, displacements maps can be reused over and over again.
6486 A description of the accepted options follows.
6490 Set displace behavior for pixels that are out of range.
6492 Available values are:
6495 Missing pixels are replaced by black pixels.
6498 Adjacent pixels will spread out to replace missing pixels.
6501 Out of range pixels are wrapped so they point to pixels of other side.
6503 Default is @samp{smear}.
6507 @subsection Examples
6511 Add ripple effect to rgb input of video size hd720:
6513 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
6517 Add wave effect to rgb input of video size hd720:
6519 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
6525 Draw a colored box on the input image.
6527 It accepts the following parameters:
6532 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6536 The expressions which specify the width and height of the box; if 0 they are interpreted as
6537 the input width and height. It defaults to 0.
6540 Specify the color of the box to write. For the general syntax of this option,
6541 check the "Color" section in the ffmpeg-utils manual. If the special
6542 value @code{invert} is used, the box edge color is the same as the
6543 video with inverted luma.
6546 The expression which sets the thickness of the box edge. Default value is @code{3}.
6548 See below for the list of accepted constants.
6551 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6552 following constants:
6556 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6560 horizontal and vertical chroma subsample values. For example for the
6561 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6565 The input width and height.
6568 The input sample aspect ratio.
6572 The x and y offset coordinates where the box is drawn.
6576 The width and height of the drawn box.
6579 The thickness of the drawn box.
6581 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6582 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6586 @subsection Examples
6590 Draw a black box around the edge of the input image:
6596 Draw a box with color red and an opacity of 50%:
6598 drawbox=10:20:200:60:red@@0.5
6601 The previous example can be specified as:
6603 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6607 Fill the box with pink color:
6609 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6613 Draw a 2-pixel red 2.40:1 mask:
6615 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
6621 Draw a grid on the input image.
6623 It accepts the following parameters:
6628 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6632 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6633 input width and height, respectively, minus @code{thickness}, so image gets
6634 framed. Default to 0.
6637 Specify the color of the grid. For the general syntax of this option,
6638 check the "Color" section in the ffmpeg-utils manual. If the special
6639 value @code{invert} is used, the grid color is the same as the
6640 video with inverted luma.
6643 The expression which sets the thickness of the grid line. Default value is @code{1}.
6645 See below for the list of accepted constants.
6648 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6649 following constants:
6653 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6657 horizontal and vertical chroma subsample values. For example for the
6658 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6662 The input grid cell width and height.
6665 The input sample aspect ratio.
6669 The x and y coordinates of some point of grid intersection (meant to configure offset).
6673 The width and height of the drawn cell.
6676 The thickness of the drawn cell.
6678 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6679 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6683 @subsection Examples
6687 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6689 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6693 Draw a white 3x3 grid with an opacity of 50%:
6695 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6702 Draw a text string or text from a specified file on top of a video, using the
6703 libfreetype library.
6705 To enable compilation of this filter, you need to configure FFmpeg with
6706 @code{--enable-libfreetype}.
6707 To enable default font fallback and the @var{font} option you need to
6708 configure FFmpeg with @code{--enable-libfontconfig}.
6709 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6710 @code{--enable-libfribidi}.
6714 It accepts the following parameters:
6719 Used to draw a box around text using the background color.
6720 The value must be either 1 (enable) or 0 (disable).
6721 The default value of @var{box} is 0.
6724 Set the width of the border to be drawn around the box using @var{boxcolor}.
6725 The default value of @var{boxborderw} is 0.
6728 The color to be used for drawing box around text. For the syntax of this
6729 option, check the "Color" section in the ffmpeg-utils manual.
6731 The default value of @var{boxcolor} is "white".
6734 Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6735 The default value of @var{line_spacing} is 0.
6738 Set the width of the border to be drawn around the text using @var{bordercolor}.
6739 The default value of @var{borderw} is 0.
6742 Set the color to be used for drawing border around text. For the syntax of this
6743 option, check the "Color" section in the ffmpeg-utils manual.
6745 The default value of @var{bordercolor} is "black".
6748 Select how the @var{text} is expanded. Can be either @code{none},
6749 @code{strftime} (deprecated) or
6750 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6754 Set a start time for the count. Value is in microseconds. Only applied
6755 in the deprecated strftime expansion mode. To emulate in normal expansion
6756 mode use the @code{pts} function, supplying the start time (in seconds)
6757 as the second argument.
6760 If true, check and fix text coords to avoid clipping.
6763 The color to be used for drawing fonts. For the syntax of this option, check
6764 the "Color" section in the ffmpeg-utils manual.
6766 The default value of @var{fontcolor} is "black".
6768 @item fontcolor_expr
6769 String which is expanded the same way as @var{text} to obtain dynamic
6770 @var{fontcolor} value. By default this option has empty value and is not
6771 processed. When this option is set, it overrides @var{fontcolor} option.
6774 The font family to be used for drawing text. By default Sans.
6777 The font file to be used for drawing text. The path must be included.
6778 This parameter is mandatory if the fontconfig support is disabled.
6781 Draw the text applying alpha blending. The value can
6782 be a number between 0.0 and 1.0.
6783 The expression accepts the same variables @var{x, y} as well.
6784 The default value is 1.
6785 Please see @var{fontcolor_expr}.
6788 The font size to be used for drawing text.
6789 The default value of @var{fontsize} is 16.
6792 If set to 1, attempt to shape the text (for example, reverse the order of
6793 right-to-left text and join Arabic characters) before drawing it.
6794 Otherwise, just draw the text exactly as given.
6795 By default 1 (if supported).
6798 The flags to be used for loading the fonts.
6800 The flags map the corresponding flags supported by libfreetype, and are
6801 a combination of the following values:
6808 @item vertical_layout
6809 @item force_autohint
6812 @item ignore_global_advance_width
6814 @item ignore_transform
6820 Default value is "default".
6822 For more information consult the documentation for the FT_LOAD_*
6826 The color to be used for drawing a shadow behind the drawn text. For the
6827 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6829 The default value of @var{shadowcolor} is "black".
6833 The x and y offsets for the text shadow position with respect to the
6834 position of the text. They can be either positive or negative
6835 values. The default value for both is "0".
6838 The starting frame number for the n/frame_num variable. The default value
6842 The size in number of spaces to use for rendering the tab.
6846 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6847 format. It can be used with or without text parameter. @var{timecode_rate}
6848 option must be specified.
6850 @item timecode_rate, rate, r
6851 Set the timecode frame rate (timecode only).
6854 If set to 1, the output of the timecode option will wrap around at 24 hours.
6855 Default is 0 (disabled).
6858 The text string to be drawn. The text must be a sequence of UTF-8
6860 This parameter is mandatory if no file is specified with the parameter
6864 A text file containing text to be drawn. The text must be a sequence
6865 of UTF-8 encoded characters.
6867 This parameter is mandatory if no text string is specified with the
6868 parameter @var{text}.
6870 If both @var{text} and @var{textfile} are specified, an error is thrown.
6873 If set to 1, the @var{textfile} will be reloaded before each frame.
6874 Be sure to update it atomically, or it may be read partially, or even fail.
6878 The expressions which specify the offsets where text will be drawn
6879 within the video frame. They are relative to the top/left border of the
6882 The default value of @var{x} and @var{y} is "0".
6884 See below for the list of accepted constants and functions.
6887 The parameters for @var{x} and @var{y} are expressions containing the
6888 following constants and functions:
6892 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6896 horizontal and vertical chroma subsample values. For example for the
6897 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6900 the height of each text line
6908 @item max_glyph_a, ascent
6909 the maximum distance from the baseline to the highest/upper grid
6910 coordinate used to place a glyph outline point, for all the rendered
6912 It is a positive value, due to the grid's orientation with the Y axis
6915 @item max_glyph_d, descent
6916 the maximum distance from the baseline to the lowest grid coordinate
6917 used to place a glyph outline point, for all the rendered glyphs.
6918 This is a negative value, due to the grid's orientation, with the Y axis
6922 maximum glyph height, that is the maximum height for all the glyphs
6923 contained in the rendered text, it is equivalent to @var{ascent} -
6927 maximum glyph width, that is the maximum width for all the glyphs
6928 contained in the rendered text
6931 the number of input frame, starting from 0
6933 @item rand(min, max)
6934 return a random number included between @var{min} and @var{max}
6937 The input sample aspect ratio.
6940 timestamp expressed in seconds, NAN if the input timestamp is unknown
6943 the height of the rendered text
6946 the width of the rendered text
6950 the x and y offset coordinates where the text is drawn.
6952 These parameters allow the @var{x} and @var{y} expressions to refer
6953 each other, so you can for example specify @code{y=x/dar}.
6956 @anchor{drawtext_expansion}
6957 @subsection Text expansion
6959 If @option{expansion} is set to @code{strftime},
6960 the filter recognizes strftime() sequences in the provided text and
6961 expands them accordingly. Check the documentation of strftime(). This
6962 feature is deprecated.
6964 If @option{expansion} is set to @code{none}, the text is printed verbatim.
6966 If @option{expansion} is set to @code{normal} (which is the default),
6967 the following expansion mechanism is used.
6969 The backslash character @samp{\}, followed by any character, always expands to
6970 the second character.
6972 Sequences of the form @code{%@{...@}} are expanded. The text between the
6973 braces is a function name, possibly followed by arguments separated by ':'.
6974 If the arguments contain special characters or delimiters (':' or '@}'),
6975 they should be escaped.
6977 Note that they probably must also be escaped as the value for the
6978 @option{text} option in the filter argument string and as the filter
6979 argument in the filtergraph description, and possibly also for the shell,
6980 that makes up to four levels of escaping; using a text file avoids these
6983 The following functions are available:
6988 The expression evaluation result.
6990 It must take one argument specifying the expression to be evaluated,
6991 which accepts the same constants and functions as the @var{x} and
6992 @var{y} values. Note that not all constants should be used, for
6993 example the text size is not known when evaluating the expression, so
6994 the constants @var{text_w} and @var{text_h} will have an undefined
6997 @item expr_int_format, eif
6998 Evaluate the expression's value and output as formatted integer.
7000 The first argument is the expression to be evaluated, just as for the @var{expr} function.
7001 The second argument specifies the output format. Allowed values are @samp{x},
7002 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7003 @code{printf} function.
7004 The third parameter is optional and sets the number of positions taken by the output.
7005 It can be used to add padding with zeros from the left.
7008 The time at which the filter is running, expressed in UTC.
7009 It can accept an argument: a strftime() format string.
7012 The time at which the filter is running, expressed in the local time zone.
7013 It can accept an argument: a strftime() format string.
7016 Frame metadata. Takes one or two arguments.
7018 The first argument is mandatory and specifies the metadata key.
7020 The second argument is optional and specifies a default value, used when the
7021 metadata key is not found or empty.
7024 The frame number, starting from 0.
7027 A 1 character description of the current picture type.
7030 The timestamp of the current frame.
7031 It can take up to three arguments.
7033 The first argument is the format of the timestamp; it defaults to @code{flt}
7034 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7035 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7036 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7037 @code{localtime} stands for the timestamp of the frame formatted as
7038 local time zone time.
7040 The second argument is an offset added to the timestamp.
7042 If the format is set to @code{localtime} or @code{gmtime},
7043 a third argument may be supplied: a strftime() format string.
7044 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7047 @subsection Examples
7051 Draw "Test Text" with font FreeSerif, using the default values for the
7052 optional parameters.
7055 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7059 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7060 and y=50 (counting from the top-left corner of the screen), text is
7061 yellow with a red box around it. Both the text and the box have an
7065 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7066 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7069 Note that the double quotes are not necessary if spaces are not used
7070 within the parameter list.
7073 Show the text at the center of the video frame:
7075 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7079 Show the text at a random position, switching to a new position every 30 seconds:
7081 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)"
7085 Show a text line sliding from right to left in the last row of the video
7086 frame. The file @file{LONG_LINE} is assumed to contain a single line
7089 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7093 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7095 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7099 Draw a single green letter "g", at the center of the input video.
7100 The glyph baseline is placed at half screen height.
7102 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7106 Show text for 1 second every 3 seconds:
7108 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7112 Use fontconfig to set the font. Note that the colons need to be escaped.
7114 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7118 Print the date of a real-time encoding (see strftime(3)):
7120 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7124 Show text fading in and out (appearing/disappearing):
7127 DS=1.0 # display start
7128 DE=10.0 # display end
7129 FID=1.5 # fade in duration
7130 FOD=5 # fade out duration
7131 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 @}"
7135 Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7136 and the @option{fontsize} value are included in the @option{y} offset.
7138 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7139 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7144 For more information about libfreetype, check:
7145 @url{http://www.freetype.org/}.
7147 For more information about fontconfig, check:
7148 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7150 For more information about libfribidi, check:
7151 @url{http://fribidi.org/}.
7155 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7157 The filter accepts the following options:
7162 Set low and high threshold values used by the Canny thresholding
7165 The high threshold selects the "strong" edge pixels, which are then
7166 connected through 8-connectivity with the "weak" edge pixels selected
7167 by the low threshold.
7169 @var{low} and @var{high} threshold values must be chosen in the range
7170 [0,1], and @var{low} should be lesser or equal to @var{high}.
7172 Default value for @var{low} is @code{20/255}, and default value for @var{high}
7176 Define the drawing mode.
7180 Draw white/gray wires on black background.
7183 Mix the colors to create a paint/cartoon effect.
7186 Default value is @var{wires}.
7189 @subsection Examples
7193 Standard edge detection with custom values for the hysteresis thresholding:
7195 edgedetect=low=0.1:high=0.4
7199 Painting effect without thresholding:
7201 edgedetect=mode=colormix:high=0
7206 Set brightness, contrast, saturation and approximate gamma adjustment.
7208 The filter accepts the following options:
7212 Set the contrast expression. The value must be a float value in range
7213 @code{-2.0} to @code{2.0}. The default value is "1".
7216 Set the brightness expression. The value must be a float value in
7217 range @code{-1.0} to @code{1.0}. The default value is "0".
7220 Set the saturation expression. The value must be a float in
7221 range @code{0.0} to @code{3.0}. The default value is "1".
7224 Set the gamma expression. The value must be a float in range
7225 @code{0.1} to @code{10.0}. The default value is "1".
7228 Set the gamma expression for red. The value must be a float in
7229 range @code{0.1} to @code{10.0}. The default value is "1".
7232 Set the gamma expression for green. The value must be a float in range
7233 @code{0.1} to @code{10.0}. The default value is "1".
7236 Set the gamma expression for blue. The value must be a float in range
7237 @code{0.1} to @code{10.0}. The default value is "1".
7240 Set the gamma weight expression. It can be used to reduce the effect
7241 of a high gamma value on bright image areas, e.g. keep them from
7242 getting overamplified and just plain white. The value must be a float
7243 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7244 gamma correction all the way down while @code{1.0} leaves it at its
7245 full strength. Default is "1".
7248 Set when the expressions for brightness, contrast, saturation and
7249 gamma expressions are evaluated.
7251 It accepts the following values:
7254 only evaluate expressions once during the filter initialization or
7255 when a command is processed
7258 evaluate expressions for each incoming frame
7261 Default value is @samp{init}.
7264 The expressions accept the following parameters:
7267 frame count of the input frame starting from 0
7270 byte position of the corresponding packet in the input file, NAN if
7274 frame rate of the input video, NAN if the input frame rate is unknown
7277 timestamp expressed in seconds, NAN if the input timestamp is unknown
7280 @subsection Commands
7281 The filter supports the following commands:
7285 Set the contrast expression.
7288 Set the brightness expression.
7291 Set the saturation expression.
7294 Set the gamma expression.
7297 Set the gamma_r expression.
7300 Set gamma_g expression.
7303 Set gamma_b expression.
7306 Set gamma_weight expression.
7308 The command accepts the same syntax of the corresponding option.
7310 If the specified expression is not valid, it is kept at its current
7317 Apply erosion effect to the video.
7319 This filter replaces the pixel by the local(3x3) minimum.
7321 It accepts the following options:
7328 Limit the maximum change for each plane, default is 65535.
7329 If 0, plane will remain unchanged.
7332 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7335 Flags to local 3x3 coordinates maps like this:
7342 @section extractplanes
7344 Extract color channel components from input video stream into
7345 separate grayscale video streams.
7347 The filter accepts the following option:
7351 Set plane(s) to extract.
7353 Available values for planes are:
7364 Choosing planes not available in the input will result in an error.
7365 That means you cannot select @code{r}, @code{g}, @code{b} planes
7366 with @code{y}, @code{u}, @code{v} planes at same time.
7369 @subsection Examples
7373 Extract luma, u and v color channel component from input video frame
7374 into 3 grayscale outputs:
7376 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
7382 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7384 For each input image, the filter will compute the optimal mapping from
7385 the input to the output given the codebook length, that is the number
7386 of distinct output colors.
7388 This filter accepts the following options.
7391 @item codebook_length, l
7392 Set codebook length. The value must be a positive integer, and
7393 represents the number of distinct output colors. Default value is 256.
7396 Set the maximum number of iterations to apply for computing the optimal
7397 mapping. The higher the value the better the result and the higher the
7398 computation time. Default value is 1.
7401 Set a random seed, must be an integer included between 0 and
7402 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7403 will try to use a good random seed on a best effort basis.
7406 Set pal8 output pixel format. This option does not work with codebook
7407 length greater than 256.
7412 Apply a fade-in/out effect to the input video.
7414 It accepts the following parameters:
7418 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7420 Default is @code{in}.
7422 @item start_frame, s
7423 Specify the number of the frame to start applying the fade
7424 effect at. Default is 0.
7427 The number of frames that the fade effect lasts. At the end of the
7428 fade-in effect, the output video will have the same intensity as the input video.
7429 At the end of the fade-out transition, the output video will be filled with the
7430 selected @option{color}.
7434 If set to 1, fade only alpha channel, if one exists on the input.
7437 @item start_time, st
7438 Specify the timestamp (in seconds) of the frame to start to apply the fade
7439 effect. If both start_frame and start_time are specified, the fade will start at
7440 whichever comes last. Default is 0.
7443 The number of seconds for which the fade effect has to last. At the end of the
7444 fade-in effect the output video will have the same intensity as the input video,
7445 at the end of the fade-out transition the output video will be filled with the
7446 selected @option{color}.
7447 If both duration and nb_frames are specified, duration is used. Default is 0
7448 (nb_frames is used by default).
7451 Specify the color of the fade. Default is "black".
7454 @subsection Examples
7458 Fade in the first 30 frames of video:
7463 The command above is equivalent to:
7469 Fade out the last 45 frames of a 200-frame video:
7472 fade=type=out:start_frame=155:nb_frames=45
7476 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7478 fade=in:0:25, fade=out:975:25
7482 Make the first 5 frames yellow, then fade in from frame 5-24:
7484 fade=in:5:20:color=yellow
7488 Fade in alpha over first 25 frames of video:
7490 fade=in:0:25:alpha=1
7494 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7496 fade=t=in:st=5.5:d=0.5
7502 Apply arbitrary expressions to samples in frequency domain
7506 Adjust the dc value (gain) of the luma plane of the image. The filter
7507 accepts an integer value in range @code{0} to @code{1000}. The default
7508 value is set to @code{0}.
7511 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7512 filter accepts an integer value in range @code{0} to @code{1000}. The
7513 default value is set to @code{0}.
7516 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7517 filter accepts an integer value in range @code{0} to @code{1000}. The
7518 default value is set to @code{0}.
7521 Set the frequency domain weight expression for the luma plane.
7524 Set the frequency domain weight expression for the 1st chroma plane.
7527 Set the frequency domain weight expression for the 2nd chroma plane.
7529 The filter accepts the following variables:
7532 The coordinates of the current sample.
7536 The width and height of the image.
7539 @subsection Examples
7545 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7551 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7557 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7563 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7570 Extract a single field from an interlaced image using stride
7571 arithmetic to avoid wasting CPU time. The output frames are marked as
7574 The filter accepts the following options:
7578 Specify whether to extract the top (if the value is @code{0} or
7579 @code{top}) or the bottom field (if the value is @code{1} or
7585 Create new frames by copying the top and bottom fields from surrounding frames
7586 supplied as numbers by the hint file.
7590 Set file containing hints: absolute/relative frame numbers.
7592 There must be one line for each frame in a clip. Each line must contain two
7593 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7594 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7595 is current frame number for @code{absolute} mode or out of [-1, 1] range
7596 for @code{relative} mode. First number tells from which frame to pick up top
7597 field and second number tells from which frame to pick up bottom field.
7599 If optionally followed by @code{+} output frame will be marked as interlaced,
7600 else if followed by @code{-} output frame will be marked as progressive, else
7601 it will be marked same as input frame.
7602 If line starts with @code{#} or @code{;} that line is skipped.
7605 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7608 Example of first several lines of @code{hint} file for @code{relative} mode:
7611 1,0 - # second frame, use third's frame top field and second's frame bottom field
7612 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7629 Field matching filter for inverse telecine. It is meant to reconstruct the
7630 progressive frames from a telecined stream. The filter does not drop duplicated
7631 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7632 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7634 The separation of the field matching and the decimation is notably motivated by
7635 the possibility of inserting a de-interlacing filter fallback between the two.
7636 If the source has mixed telecined and real interlaced content,
7637 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7638 But these remaining combed frames will be marked as interlaced, and thus can be
7639 de-interlaced by a later filter such as @ref{yadif} before decimation.
7641 In addition to the various configuration options, @code{fieldmatch} can take an
7642 optional second stream, activated through the @option{ppsrc} option. If
7643 enabled, the frames reconstruction will be based on the fields and frames from
7644 this second stream. This allows the first input to be pre-processed in order to
7645 help the various algorithms of the filter, while keeping the output lossless
7646 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7647 or brightness/contrast adjustments can help.
7649 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7650 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7651 which @code{fieldmatch} is based on. While the semantic and usage are very
7652 close, some behaviour and options names can differ.
7654 The @ref{decimate} filter currently only works for constant frame rate input.
7655 If your input has mixed telecined (30fps) and progressive content with a lower
7656 framerate like 24fps use the following filterchain to produce the necessary cfr
7657 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7659 The filter accepts the following options:
7663 Specify the assumed field order of the input stream. Available values are:
7667 Auto detect parity (use FFmpeg's internal parity value).
7669 Assume bottom field first.
7671 Assume top field first.
7674 Note that it is sometimes recommended not to trust the parity announced by the
7677 Default value is @var{auto}.
7680 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7681 sense that it won't risk creating jerkiness due to duplicate frames when
7682 possible, but if there are bad edits or blended fields it will end up
7683 outputting combed frames when a good match might actually exist. On the other
7684 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7685 but will almost always find a good frame if there is one. The other values are
7686 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7687 jerkiness and creating duplicate frames versus finding good matches in sections
7688 with bad edits, orphaned fields, blended fields, etc.
7690 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7692 Available values are:
7696 2-way matching (p/c)
7698 2-way matching, and trying 3rd match if still combed (p/c + n)
7700 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7702 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7703 still combed (p/c + n + u/b)
7705 3-way matching (p/c/n)
7707 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7708 detected as combed (p/c/n + u/b)
7711 The parenthesis at the end indicate the matches that would be used for that
7712 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7715 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7718 Default value is @var{pc_n}.
7721 Mark the main input stream as a pre-processed input, and enable the secondary
7722 input stream as the clean source to pick the fields from. See the filter
7723 introduction for more details. It is similar to the @option{clip2} feature from
7726 Default value is @code{0} (disabled).
7729 Set the field to match from. It is recommended to set this to the same value as
7730 @option{order} unless you experience matching failures with that setting. In
7731 certain circumstances changing the field that is used to match from can have a
7732 large impact on matching performance. Available values are:
7736 Automatic (same value as @option{order}).
7738 Match from the bottom field.
7740 Match from the top field.
7743 Default value is @var{auto}.
7746 Set whether or not chroma is included during the match comparisons. In most
7747 cases it is recommended to leave this enabled. You should set this to @code{0}
7748 only if your clip has bad chroma problems such as heavy rainbowing or other
7749 artifacts. Setting this to @code{0} could also be used to speed things up at
7750 the cost of some accuracy.
7752 Default value is @code{1}.
7756 These define an exclusion band which excludes the lines between @option{y0} and
7757 @option{y1} from being included in the field matching decision. An exclusion
7758 band can be used to ignore subtitles, a logo, or other things that may
7759 interfere with the matching. @option{y0} sets the starting scan line and
7760 @option{y1} sets the ending line; all lines in between @option{y0} and
7761 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7762 @option{y0} and @option{y1} to the same value will disable the feature.
7763 @option{y0} and @option{y1} defaults to @code{0}.
7766 Set the scene change detection threshold as a percentage of maximum change on
7767 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7768 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7769 @option{scthresh} is @code{[0.0, 100.0]}.
7771 Default value is @code{12.0}.
7774 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7775 account the combed scores of matches when deciding what match to use as the
7776 final match. Available values are:
7780 No final matching based on combed scores.
7782 Combed scores are only used when a scene change is detected.
7784 Use combed scores all the time.
7787 Default is @var{sc}.
7790 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7791 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7792 Available values are:
7796 No forced calculation.
7798 Force p/c/n calculations.
7800 Force p/c/n/u/b calculations.
7803 Default value is @var{none}.
7806 This is the area combing threshold used for combed frame detection. This
7807 essentially controls how "strong" or "visible" combing must be to be detected.
7808 Larger values mean combing must be more visible and smaller values mean combing
7809 can be less visible or strong and still be detected. Valid settings are from
7810 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7811 be detected as combed). This is basically a pixel difference value. A good
7812 range is @code{[8, 12]}.
7814 Default value is @code{9}.
7817 Sets whether or not chroma is considered in the combed frame decision. Only
7818 disable this if your source has chroma problems (rainbowing, etc.) that are
7819 causing problems for the combed frame detection with chroma enabled. Actually,
7820 using @option{chroma}=@var{0} is usually more reliable, except for the case
7821 where there is chroma only combing in the source.
7823 Default value is @code{0}.
7827 Respectively set the x-axis and y-axis size of the window used during combed
7828 frame detection. This has to do with the size of the area in which
7829 @option{combpel} pixels are required to be detected as combed for a frame to be
7830 declared combed. See the @option{combpel} parameter description for more info.
7831 Possible values are any number that is a power of 2 starting at 4 and going up
7834 Default value is @code{16}.
7837 The number of combed pixels inside any of the @option{blocky} by
7838 @option{blockx} size blocks on the frame for the frame to be detected as
7839 combed. While @option{cthresh} controls how "visible" the combing must be, this
7840 setting controls "how much" combing there must be in any localized area (a
7841 window defined by the @option{blockx} and @option{blocky} settings) on the
7842 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7843 which point no frames will ever be detected as combed). This setting is known
7844 as @option{MI} in TFM/VFM vocabulary.
7846 Default value is @code{80}.
7849 @anchor{p/c/n/u/b meaning}
7850 @subsection p/c/n/u/b meaning
7852 @subsubsection p/c/n
7854 We assume the following telecined stream:
7857 Top fields: 1 2 2 3 4
7858 Bottom fields: 1 2 3 4 4
7861 The numbers correspond to the progressive frame the fields relate to. Here, the
7862 first two frames are progressive, the 3rd and 4th are combed, and so on.
7864 When @code{fieldmatch} is configured to run a matching from bottom
7865 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7870 B 1 2 3 4 4 <-- matching reference
7879 As a result of the field matching, we can see that some frames get duplicated.
7880 To perform a complete inverse telecine, you need to rely on a decimation filter
7881 after this operation. See for instance the @ref{decimate} filter.
7883 The same operation now matching from top fields (@option{field}=@var{top})
7888 T 1 2 2 3 4 <-- matching reference
7898 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7899 basically, they refer to the frame and field of the opposite parity:
7902 @item @var{p} matches the field of the opposite parity in the previous frame
7903 @item @var{c} matches the field of the opposite parity in the current frame
7904 @item @var{n} matches the field of the opposite parity in the next frame
7909 The @var{u} and @var{b} matching are a bit special in the sense that they match
7910 from the opposite parity flag. In the following examples, we assume that we are
7911 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7912 'x' is placed above and below each matched fields.
7914 With bottom matching (@option{field}=@var{bottom}):
7919 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7920 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7928 With top matching (@option{field}=@var{top}):
7933 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7934 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7942 @subsection Examples
7944 Simple IVTC of a top field first telecined stream:
7946 fieldmatch=order=tff:combmatch=none, decimate
7949 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7951 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7956 Transform the field order of the input video.
7958 It accepts the following parameters:
7963 The output field order. Valid values are @var{tff} for top field first or @var{bff}
7964 for bottom field first.
7967 The default value is @samp{tff}.
7969 The transformation is done by shifting the picture content up or down
7970 by one line, and filling the remaining line with appropriate picture content.
7971 This method is consistent with most broadcast field order converters.
7973 If the input video is not flagged as being interlaced, or it is already
7974 flagged as being of the required output field order, then this filter does
7975 not alter the incoming video.
7977 It is very useful when converting to or from PAL DV material,
7978 which is bottom field first.
7982 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
7985 @section fifo, afifo
7987 Buffer input images and send them when they are requested.
7989 It is mainly useful when auto-inserted by the libavfilter
7992 It does not take parameters.
7996 Find a rectangular object
7998 It accepts the following options:
8002 Filepath of the object image, needs to be in gray8.
8005 Detection threshold, default is 0.5.
8008 Number of mipmaps, default is 3.
8010 @item xmin, ymin, xmax, ymax
8011 Specifies the rectangle in which to search.
8014 @subsection Examples
8018 Generate a representative palette of a given video using @command{ffmpeg}:
8020 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8026 Cover a rectangular object
8028 It accepts the following options:
8032 Filepath of the optional cover image, needs to be in yuv420.
8037 It accepts the following values:
8040 cover it by the supplied image
8042 cover it by interpolating the surrounding pixels
8045 Default value is @var{blur}.
8048 @subsection Examples
8052 Generate a representative palette of a given video using @command{ffmpeg}:
8054 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8061 Convert the input video to one of the specified pixel formats.
8062 Libavfilter will try to pick one that is suitable as input to
8065 It accepts the following parameters:
8069 A '|'-separated list of pixel format names, such as
8070 "pix_fmts=yuv420p|monow|rgb24".
8074 @subsection Examples
8078 Convert the input video to the @var{yuv420p} format
8080 format=pix_fmts=yuv420p
8083 Convert the input video to any of the formats in the list
8085 format=pix_fmts=yuv420p|yuv444p|yuv410p
8092 Convert the video to specified constant frame rate by duplicating or dropping
8093 frames as necessary.
8095 It accepts the following parameters:
8099 The desired output frame rate. The default is @code{25}.
8104 Possible values are:
8107 zero round towards 0
8111 round towards -infinity
8113 round towards +infinity
8117 The default is @code{near}.
8120 Assume the first PTS should be the given value, in seconds. This allows for
8121 padding/trimming at the start of stream. By default, no assumption is made
8122 about the first frame's expected PTS, so no padding or trimming is done.
8123 For example, this could be set to 0 to pad the beginning with duplicates of
8124 the first frame if a video stream starts after the audio stream or to trim any
8125 frames with a negative PTS.
8129 Alternatively, the options can be specified as a flat string:
8130 @var{fps}[:@var{round}].
8132 See also the @ref{setpts} filter.
8134 @subsection Examples
8138 A typical usage in order to set the fps to 25:
8144 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8146 fps=fps=film:round=near
8152 Pack two different video streams into a stereoscopic video, setting proper
8153 metadata on supported codecs. The two views should have the same size and
8154 framerate and processing will stop when the shorter video ends. Please note
8155 that you may conveniently adjust view properties with the @ref{scale} and
8158 It accepts the following parameters:
8162 The desired packing format. Supported values are:
8167 The views are next to each other (default).
8170 The views are on top of each other.
8173 The views are packed by line.
8176 The views are packed by column.
8179 The views are temporally interleaved.
8188 # Convert left and right views into a frame-sequential video
8189 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8191 # Convert views into a side-by-side video with the same output resolution as the input
8192 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
8197 Change the frame rate by interpolating new video output frames from the source
8200 This filter is not designed to function correctly with interlaced media. If
8201 you wish to change the frame rate of interlaced media then you are required
8202 to deinterlace before this filter and re-interlace after this filter.
8204 A description of the accepted options follows.
8208 Specify the output frames per second. This option can also be specified
8209 as a value alone. The default is @code{50}.
8212 Specify the start of a range where the output frame will be created as a
8213 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8214 the default is @code{15}.
8217 Specify the end of a range where the output frame will be created as a
8218 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8219 the default is @code{240}.
8222 Specify the level at which a scene change is detected as a value between
8223 0 and 100 to indicate a new scene; a low value reflects a low
8224 probability for the current frame to introduce a new scene, while a higher
8225 value means the current frame is more likely to be one.
8226 The default is @code{7}.
8229 Specify flags influencing the filter process.
8231 Available value for @var{flags} is:
8234 @item scene_change_detect, scd
8235 Enable scene change detection using the value of the option @var{scene}.
8236 This flag is enabled by default.
8242 Select one frame every N-th frame.
8244 This filter accepts the following option:
8247 Select frame after every @code{step} frames.
8248 Allowed values are positive integers higher than 0. Default value is @code{1}.
8254 Apply a frei0r effect to the input video.
8256 To enable the compilation of this filter, you need to install the frei0r
8257 header and configure FFmpeg with @code{--enable-frei0r}.
8259 It accepts the following parameters:
8264 The name of the frei0r effect to load. If the environment variable
8265 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8266 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8267 Otherwise, the standard frei0r paths are searched, in this order:
8268 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8269 @file{/usr/lib/frei0r-1/}.
8272 A '|'-separated list of parameters to pass to the frei0r effect.
8276 A frei0r effect parameter can be a boolean (its value is either
8277 "y" or "n"), a double, a color (specified as
8278 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8279 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8280 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8281 @var{X} and @var{Y} are floating point numbers) and/or a string.
8283 The number and types of parameters depend on the loaded effect. If an
8284 effect parameter is not specified, the default value is set.
8286 @subsection Examples
8290 Apply the distort0r effect, setting the first two double parameters:
8292 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8296 Apply the colordistance effect, taking a color as the first parameter:
8298 frei0r=colordistance:0.2/0.3/0.4
8299 frei0r=colordistance:violet
8300 frei0r=colordistance:0x112233
8304 Apply the perspective effect, specifying the top left and top right image
8307 frei0r=perspective:0.2/0.2|0.8/0.2
8311 For more information, see
8312 @url{http://frei0r.dyne.org}
8316 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8318 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8319 processing filter, one of them is performed once per block, not per pixel.
8320 This allows for much higher speed.
8322 The filter accepts the following options:
8326 Set quality. This option defines the number of levels for averaging. It accepts
8327 an integer in the range 4-5. Default value is @code{4}.
8330 Force a constant quantization parameter. It accepts an integer in range 0-63.
8331 If not set, the filter will use the QP from the video stream (if available).
8334 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8335 more details but also more artifacts, while higher values make the image smoother
8336 but also blurrier. Default value is @code{0} − PSNR optimal.
8339 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8340 option may cause flicker since the B-Frames have often larger QP. Default is
8341 @code{0} (not enabled).
8347 Apply Gaussian blur filter.
8349 The filter accepts the following options:
8353 Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8356 Set number of steps for Gaussian approximation. Defauls is @code{1}.
8359 Set which planes to filter. By default all planes are filtered.
8362 Set vertical sigma, if negative it will be same as @code{sigma}.
8363 Default is @code{-1}.
8368 The filter accepts the following options:
8372 Set the luminance expression.
8374 Set the chrominance blue expression.
8376 Set the chrominance red expression.
8378 Set the alpha expression.
8380 Set the red expression.
8382 Set the green expression.
8384 Set the blue expression.
8387 The colorspace is selected according to the specified options. If one
8388 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8389 options is specified, the filter will automatically select a YCbCr
8390 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8391 @option{blue_expr} options is specified, it will select an RGB
8394 If one of the chrominance expression is not defined, it falls back on the other
8395 one. If no alpha expression is specified it will evaluate to opaque value.
8396 If none of chrominance expressions are specified, they will evaluate
8397 to the luminance expression.
8399 The expressions can use the following variables and functions:
8403 The sequential number of the filtered frame, starting from @code{0}.
8407 The coordinates of the current sample.
8411 The width and height of the image.
8415 Width and height scale depending on the currently filtered plane. It is the
8416 ratio between the corresponding luma plane number of pixels and the current
8417 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8418 @code{0.5,0.5} for chroma planes.
8421 Time of the current frame, expressed in seconds.
8424 Return the value of the pixel at location (@var{x},@var{y}) of the current
8428 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8432 Return the value of the pixel at location (@var{x},@var{y}) of the
8433 blue-difference chroma plane. Return 0 if there is no such plane.
8436 Return the value of the pixel at location (@var{x},@var{y}) of the
8437 red-difference chroma plane. Return 0 if there is no such plane.
8442 Return the value of the pixel at location (@var{x},@var{y}) of the
8443 red/green/blue component. Return 0 if there is no such component.
8446 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8447 plane. Return 0 if there is no such plane.
8450 For functions, if @var{x} and @var{y} are outside the area, the value will be
8451 automatically clipped to the closer edge.
8453 @subsection Examples
8457 Flip the image horizontally:
8463 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8464 wavelength of 100 pixels:
8466 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8470 Generate a fancy enigmatic moving light:
8472 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
8476 Generate a quick emboss effect:
8478 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8482 Modify RGB components depending on pixel position:
8484 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8488 Create a radial gradient that is the same size as the input (also see
8489 the @ref{vignette} filter):
8491 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8497 Fix the banding artifacts that are sometimes introduced into nearly flat
8498 regions by truncation to 8-bit color depth.
8499 Interpolate the gradients that should go where the bands are, and
8502 It is designed for playback only. Do not use it prior to
8503 lossy compression, because compression tends to lose the dither and
8504 bring back the bands.
8506 It accepts the following parameters:
8511 The maximum amount by which the filter will change any one pixel. This is also
8512 the threshold for detecting nearly flat regions. Acceptable values range from
8513 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8517 The neighborhood to fit the gradient to. A larger radius makes for smoother
8518 gradients, but also prevents the filter from modifying the pixels near detailed
8519 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8520 values will be clipped to the valid range.
8524 Alternatively, the options can be specified as a flat string:
8525 @var{strength}[:@var{radius}]
8527 @subsection Examples
8531 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8537 Specify radius, omitting the strength (which will fall-back to the default
8548 Apply a Hald CLUT to a video stream.
8550 First input is the video stream to process, and second one is the Hald CLUT.
8551 The Hald CLUT input can be a simple picture or a complete video stream.
8553 The filter accepts the following options:
8557 Force termination when the shortest input terminates. Default is @code{0}.
8559 Continue applying the last CLUT after the end of the stream. A value of
8560 @code{0} disable the filter after the last frame of the CLUT is reached.
8561 Default is @code{1}.
8564 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8565 filters share the same internals).
8567 More information about the Hald CLUT can be found on Eskil Steenberg's website
8568 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8570 @subsection Workflow examples
8572 @subsubsection Hald CLUT video stream
8574 Generate an identity Hald CLUT stream altered with various effects:
8576 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
8579 Note: make sure you use a lossless codec.
8581 Then use it with @code{haldclut} to apply it on some random stream:
8583 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8586 The Hald CLUT will be applied to the 10 first seconds (duration of
8587 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8588 to the remaining frames of the @code{mandelbrot} stream.
8590 @subsubsection Hald CLUT with preview
8592 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8593 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8594 biggest possible square starting at the top left of the picture. The remaining
8595 padding pixels (bottom or right) will be ignored. This area can be used to add
8596 a preview of the Hald CLUT.
8598 Typically, the following generated Hald CLUT will be supported by the
8599 @code{haldclut} filter:
8602 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8603 pad=iw+320 [padded_clut];
8604 smptebars=s=320x256, split [a][b];
8605 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8606 [main][b] overlay=W-320" -frames:v 1 clut.png
8609 It contains the original and a preview of the effect of the CLUT: SMPTE color
8610 bars are displayed on the right-top, and below the same color bars processed by
8613 Then, the effect of this Hald CLUT can be visualized with:
8615 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8620 Flip the input video horizontally.
8622 For example, to horizontally flip the input video with @command{ffmpeg}:
8624 ffmpeg -i in.avi -vf "hflip" out.avi
8628 This filter applies a global color histogram equalization on a
8631 It can be used to correct video that has a compressed range of pixel
8632 intensities. The filter redistributes the pixel intensities to
8633 equalize their distribution across the intensity range. It may be
8634 viewed as an "automatically adjusting contrast filter". This filter is
8635 useful only for correcting degraded or poorly captured source
8638 The filter accepts the following options:
8642 Determine the amount of equalization to be applied. As the strength
8643 is reduced, the distribution of pixel intensities more-and-more
8644 approaches that of the input frame. The value must be a float number
8645 in the range [0,1] and defaults to 0.200.
8648 Set the maximum intensity that can generated and scale the output
8649 values appropriately. The strength should be set as desired and then
8650 the intensity can be limited if needed to avoid washing-out. The value
8651 must be a float number in the range [0,1] and defaults to 0.210.
8654 Set the antibanding level. If enabled the filter will randomly vary
8655 the luminance of output pixels by a small amount to avoid banding of
8656 the histogram. Possible values are @code{none}, @code{weak} or
8657 @code{strong}. It defaults to @code{none}.
8662 Compute and draw a color distribution histogram for the input video.
8664 The computed histogram is a representation of the color component
8665 distribution in an image.
8667 Standard histogram displays the color components distribution in an image.
8668 Displays color graph for each color component. Shows distribution of
8669 the Y, U, V, A or R, G, B components, depending on input format, in the
8670 current frame. Below each graph a color component scale meter is shown.
8672 The filter accepts the following options:
8676 Set height of level. Default value is @code{200}.
8677 Allowed range is [50, 2048].
8680 Set height of color scale. Default value is @code{12}.
8681 Allowed range is [0, 40].
8685 It accepts the following values:
8688 Per color component graphs are placed below each other.
8691 Presents information identical to that in the @code{parade}, except
8692 that the graphs representing color components are superimposed directly
8695 Default is @code{parade}.
8698 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8699 Default is @code{linear}.
8702 Set what color components to display.
8703 Default is @code{7}.
8706 Set foreground opacity. Default is @code{0.7}.
8709 Set background opacity. Default is @code{0.5}.
8712 @subsection Examples
8717 Calculate and draw histogram:
8719 ffplay -i input -vf histogram
8727 This is a high precision/quality 3d denoise filter. It aims to reduce
8728 image noise, producing smooth images and making still images really
8729 still. It should enhance compressibility.
8731 It accepts the following optional parameters:
8735 A non-negative floating point number which specifies spatial luma strength.
8738 @item chroma_spatial
8739 A non-negative floating point number which specifies spatial chroma strength.
8740 It defaults to 3.0*@var{luma_spatial}/4.0.
8743 A floating point number which specifies luma temporal strength. It defaults to
8744 6.0*@var{luma_spatial}/4.0.
8747 A floating point number which specifies chroma temporal strength. It defaults to
8748 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8751 @anchor{hwupload_cuda}
8752 @section hwupload_cuda
8754 Upload system memory frames to a CUDA device.
8756 It accepts the following optional parameters:
8760 The number of the CUDA device to use
8765 Apply a high-quality magnification filter designed for pixel art. This filter
8766 was originally created by Maxim Stepin.
8768 It accepts the following option:
8772 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8773 @code{hq3x} and @code{4} for @code{hq4x}.
8774 Default is @code{3}.
8778 Stack input videos horizontally.
8780 All streams must be of same pixel format and of same height.
8782 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8783 to create same output.
8785 The filter accept the following option:
8789 Set number of input streams. Default is 2.
8792 If set to 1, force the output to terminate when the shortest input
8793 terminates. Default value is 0.
8798 Modify the hue and/or the saturation of the input.
8800 It accepts the following parameters:
8804 Specify the hue angle as a number of degrees. It accepts an expression,
8805 and defaults to "0".
8808 Specify the saturation in the [-10,10] range. It accepts an expression and
8812 Specify the hue angle as a number of radians. It accepts an
8813 expression, and defaults to "0".
8816 Specify the brightness in the [-10,10] range. It accepts an expression and
8820 @option{h} and @option{H} are mutually exclusive, and can't be
8821 specified at the same time.
8823 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8824 expressions containing the following constants:
8828 frame count of the input frame starting from 0
8831 presentation timestamp of the input frame expressed in time base units
8834 frame rate of the input video, NAN if the input frame rate is unknown
8837 timestamp expressed in seconds, NAN if the input timestamp is unknown
8840 time base of the input video
8843 @subsection Examples
8847 Set the hue to 90 degrees and the saturation to 1.0:
8853 Same command but expressing the hue in radians:
8859 Rotate hue and make the saturation swing between 0
8860 and 2 over a period of 1 second:
8862 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8866 Apply a 3 seconds saturation fade-in effect starting at 0:
8871 The general fade-in expression can be written as:
8873 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8877 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8879 hue="s=max(0\, min(1\, (8-t)/3))"
8882 The general fade-out expression can be written as:
8884 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8889 @subsection Commands
8891 This filter supports the following commands:
8897 Modify the hue and/or the saturation and/or brightness of the input video.
8898 The command accepts the same syntax of the corresponding option.
8900 If the specified expression is not valid, it is kept at its current
8906 Grow first stream into second stream by connecting components.
8907 This makes it possible to build more robust edge masks.
8909 This filter accepts the following options:
8913 Set which planes will be processed as bitmap, unprocessed planes will be
8914 copied from first stream.
8915 By default value 0xf, all planes will be processed.
8918 Set threshold which is used in filtering. If pixel component value is higher than
8919 this value filter algorithm for connecting components is activated.
8920 By default value is 0.
8925 Detect video interlacing type.
8927 This filter tries to detect if the input frames are interlaced, progressive,
8928 top or bottom field first. It will also try to detect fields that are
8929 repeated between adjacent frames (a sign of telecine).
8931 Single frame detection considers only immediately adjacent frames when classifying each frame.
8932 Multiple frame detection incorporates the classification history of previous frames.
8934 The filter will log these metadata values:
8937 @item single.current_frame
8938 Detected type of current frame using single-frame detection. One of:
8939 ``tff'' (top field first), ``bff'' (bottom field first),
8940 ``progressive'', or ``undetermined''
8943 Cumulative number of frames detected as top field first using single-frame detection.
8946 Cumulative number of frames detected as top field first using multiple-frame detection.
8949 Cumulative number of frames detected as bottom field first using single-frame detection.
8951 @item multiple.current_frame
8952 Detected type of current frame using multiple-frame detection. One of:
8953 ``tff'' (top field first), ``bff'' (bottom field first),
8954 ``progressive'', or ``undetermined''
8957 Cumulative number of frames detected as bottom field first using multiple-frame detection.
8959 @item single.progressive
8960 Cumulative number of frames detected as progressive using single-frame detection.
8962 @item multiple.progressive
8963 Cumulative number of frames detected as progressive using multiple-frame detection.
8965 @item single.undetermined
8966 Cumulative number of frames that could not be classified using single-frame detection.
8968 @item multiple.undetermined
8969 Cumulative number of frames that could not be classified using multiple-frame detection.
8971 @item repeated.current_frame
8972 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
8974 @item repeated.neither
8975 Cumulative number of frames with no repeated field.
8978 Cumulative number of frames with the top field repeated from the previous frame's top field.
8980 @item repeated.bottom
8981 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
8984 The filter accepts the following options:
8988 Set interlacing threshold.
8990 Set progressive threshold.
8992 Threshold for repeated field detection.
8994 Number of frames after which a given frame's contribution to the
8995 statistics is halved (i.e., it contributes only 0.5 to its
8996 classification). The default of 0 means that all frames seen are given
8997 full weight of 1.0 forever.
8998 @item analyze_interlaced_flag
8999 When this is not 0 then idet will use the specified number of frames to determine
9000 if the interlaced flag is accurate, it will not count undetermined frames.
9001 If the flag is found to be accurate it will be used without any further
9002 computations, if it is found to be inaccurate it will be cleared without any
9003 further computations. This allows inserting the idet filter as a low computational
9004 method to clean up the interlaced flag
9009 Deinterleave or interleave fields.
9011 This filter allows one to process interlaced images fields without
9012 deinterlacing them. Deinterleaving splits the input frame into 2
9013 fields (so called half pictures). Odd lines are moved to the top
9014 half of the output image, even lines to the bottom half.
9015 You can process (filter) them independently and then re-interleave them.
9017 The filter accepts the following options:
9021 @item chroma_mode, c
9023 Available values for @var{luma_mode}, @var{chroma_mode} and
9024 @var{alpha_mode} are:
9030 @item deinterleave, d
9031 Deinterleave fields, placing one above the other.
9034 Interleave fields. Reverse the effect of deinterleaving.
9036 Default value is @code{none}.
9039 @item chroma_swap, cs
9040 @item alpha_swap, as
9041 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9046 Apply inflate effect to the video.
9048 This filter replaces the pixel by the local(3x3) average by taking into account
9049 only values higher than the pixel.
9051 It accepts the following options:
9058 Limit the maximum change for each plane, default is 65535.
9059 If 0, plane will remain unchanged.
9064 Simple interlacing filter from progressive contents. This interleaves upper (or
9065 lower) lines from odd frames with lower (or upper) lines from even frames,
9066 halving the frame rate and preserving image height.
9069 Original Original New Frame
9070 Frame 'j' Frame 'j+1' (tff)
9071 ========== =========== ==================
9072 Line 0 --------------------> Frame 'j' Line 0
9073 Line 1 Line 1 ----> Frame 'j+1' Line 1
9074 Line 2 ---------------------> Frame 'j' Line 2
9075 Line 3 Line 3 ----> Frame 'j+1' Line 3
9077 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9080 It accepts the following optional parameters:
9084 This determines whether the interlaced frame is taken from the even
9085 (tff - default) or odd (bff) lines of the progressive frame.
9088 Enable (default) or disable the vertical lowpass filter to avoid twitter
9089 interlacing and reduce moire patterns.
9094 Deinterlace input video by applying Donald Graft's adaptive kernel
9095 deinterling. Work on interlaced parts of a video to produce
9098 The description of the accepted parameters follows.
9102 Set the threshold which affects the filter's tolerance when
9103 determining if a pixel line must be processed. It must be an integer
9104 in the range [0,255] and defaults to 10. A value of 0 will result in
9105 applying the process on every pixels.
9108 Paint pixels exceeding the threshold value to white if set to 1.
9112 Set the fields order. Swap fields if set to 1, leave fields alone if
9116 Enable additional sharpening if set to 1. Default is 0.
9119 Enable twoway sharpening if set to 1. Default is 0.
9122 @subsection Examples
9126 Apply default values:
9128 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9132 Enable additional sharpening:
9138 Paint processed pixels in white:
9144 @section lenscorrection
9146 Correct radial lens distortion
9148 This filter can be used to correct for radial distortion as can result from the use
9149 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9150 one can use tools available for example as part of opencv or simply trial-and-error.
9151 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9152 and extract the k1 and k2 coefficients from the resulting matrix.
9154 Note that effectively the same filter is available in the open-source tools Krita and
9155 Digikam from the KDE project.
9157 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9158 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9159 brightness distribution, so you may want to use both filters together in certain
9160 cases, though you will have to take care of ordering, i.e. whether vignetting should
9161 be applied before or after lens correction.
9165 The filter accepts the following options:
9169 Relative x-coordinate of the focal point of the image, and thereby the center of the
9170 distortion. This value has a range [0,1] and is expressed as fractions of the image
9173 Relative y-coordinate of the focal point of the image, and thereby the center of the
9174 distortion. This value has a range [0,1] and is expressed as fractions of the image
9177 Coefficient of the quadratic correction term. 0.5 means no correction.
9179 Coefficient of the double quadratic correction term. 0.5 means no correction.
9182 The formula that generates the correction is:
9184 @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)
9186 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9187 distances from the focal point in the source and target images, respectively.
9193 The filter accepts the following options:
9197 Set the number of loops.
9200 Set maximal size in number of frames.
9203 Set first frame of loop.
9209 Apply a 3D LUT to an input video.
9211 The filter accepts the following options:
9215 Set the 3D LUT file name.
9217 Currently supported formats:
9229 Select interpolation mode.
9231 Available values are:
9235 Use values from the nearest defined point.
9237 Interpolate values using the 8 points defining a cube.
9239 Interpolate values using a tetrahedron.
9243 @section lut, lutrgb, lutyuv
9245 Compute a look-up table for binding each pixel component input value
9246 to an output value, and apply it to the input video.
9248 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9249 to an RGB input video.
9251 These filters accept the following parameters:
9254 set first pixel component expression
9256 set second pixel component expression
9258 set third pixel component expression
9260 set fourth pixel component expression, corresponds to the alpha component
9263 set red component expression
9265 set green component expression
9267 set blue component expression
9269 alpha component expression
9272 set Y/luminance component expression
9274 set U/Cb component expression
9276 set V/Cr component expression
9279 Each of them specifies the expression to use for computing the lookup table for
9280 the corresponding pixel component values.
9282 The exact component associated to each of the @var{c*} options depends on the
9285 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9286 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9288 The expressions can contain the following constants and functions:
9293 The input width and height.
9296 The input value for the pixel component.
9299 The input value, clipped to the @var{minval}-@var{maxval} range.
9302 The maximum value for the pixel component.
9305 The minimum value for the pixel component.
9308 The negated value for the pixel component value, clipped to the
9309 @var{minval}-@var{maxval} range; it corresponds to the expression
9310 "maxval-clipval+minval".
9313 The computed value in @var{val}, clipped to the
9314 @var{minval}-@var{maxval} range.
9316 @item gammaval(gamma)
9317 The computed gamma correction value of the pixel component value,
9318 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9320 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9324 All expressions default to "val".
9326 @subsection Examples
9332 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9333 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9336 The above is the same as:
9338 lutrgb="r=negval:g=negval:b=negval"
9339 lutyuv="y=negval:u=negval:v=negval"
9349 Remove chroma components, turning the video into a graytone image:
9351 lutyuv="u=128:v=128"
9355 Apply a luma burning effect:
9361 Remove green and blue components:
9367 Set a constant alpha channel value on input:
9369 format=rgba,lutrgb=a="maxval-minval/2"
9373 Correct luminance gamma by a factor of 0.5:
9375 lutyuv=y=gammaval(0.5)
9379 Discard least significant bits of luma:
9381 lutyuv=y='bitand(val, 128+64+32)'
9385 Technicolor like effect:
9387 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9393 Compute and apply a lookup table from two video inputs.
9395 This filter accepts the following parameters:
9398 set first pixel component expression
9400 set second pixel component expression
9402 set third pixel component expression
9404 set fourth pixel component expression, corresponds to the alpha component
9407 Each of them specifies the expression to use for computing the lookup table for
9408 the corresponding pixel component values.
9410 The exact component associated to each of the @var{c*} options depends on the
9413 The expressions can contain the following constants:
9418 The input width and height.
9421 The first input value for the pixel component.
9424 The second input value for the pixel component.
9427 The first input video bit depth.
9430 The second input video bit depth.
9433 All expressions default to "x".
9435 @subsection Examples
9439 Highlight differences between two RGB video streams:
9441 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)'
9445 Highlight differences between two YUV video streams:
9447 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)'
9451 @section maskedclamp
9453 Clamp the first input stream with the second input and third input stream.
9455 Returns the value of first stream to be between second input
9456 stream - @code{undershoot} and third input stream + @code{overshoot}.
9458 This filter accepts the following options:
9461 Default value is @code{0}.
9464 Default value is @code{0}.
9467 Set which planes will be processed as bitmap, unprocessed planes will be
9468 copied from first stream.
9469 By default value 0xf, all planes will be processed.
9472 @section maskedmerge
9474 Merge the first input stream with the second input stream using per pixel
9475 weights in the third input stream.
9477 A value of 0 in the third stream pixel component means that pixel component
9478 from first stream is returned unchanged, while maximum value (eg. 255 for
9479 8-bit videos) means that pixel component from second stream is returned
9480 unchanged. Intermediate values define the amount of merging between both
9481 input stream's pixel components.
9483 This filter accepts the following options:
9486 Set which planes will be processed as bitmap, unprocessed planes will be
9487 copied from first stream.
9488 By default value 0xf, all planes will be processed.
9493 Apply motion-compensation deinterlacing.
9495 It needs one field per frame as input and must thus be used together
9496 with yadif=1/3 or equivalent.
9498 This filter accepts the following options:
9501 Set the deinterlacing mode.
9503 It accepts one of the following values:
9508 use iterative motion estimation
9510 like @samp{slow}, but use multiple reference frames.
9512 Default value is @samp{fast}.
9515 Set the picture field parity assumed for the input video. It must be
9516 one of the following values:
9520 assume top field first
9522 assume bottom field first
9525 Default value is @samp{bff}.
9528 Set per-block quantization parameter (QP) used by the internal
9531 Higher values should result in a smoother motion vector field but less
9532 optimal individual vectors. Default value is 1.
9535 @section mergeplanes
9537 Merge color channel components from several video streams.
9539 The filter accepts up to 4 input streams, and merge selected input
9540 planes to the output video.
9542 This filter accepts the following options:
9545 Set input to output plane mapping. Default is @code{0}.
9547 The mappings is specified as a bitmap. It should be specified as a
9548 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9549 mapping for the first plane of the output stream. 'A' sets the number of
9550 the input stream to use (from 0 to 3), and 'a' the plane number of the
9551 corresponding input to use (from 0 to 3). The rest of the mappings is
9552 similar, 'Bb' describes the mapping for the output stream second
9553 plane, 'Cc' describes the mapping for the output stream third plane and
9554 'Dd' describes the mapping for the output stream fourth plane.
9557 Set output pixel format. Default is @code{yuva444p}.
9560 @subsection Examples
9564 Merge three gray video streams of same width and height into single video stream:
9566 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9570 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9572 [a0][a1]mergeplanes=0x00010210:yuva444p
9576 Swap Y and A plane in yuva444p stream:
9578 format=yuva444p,mergeplanes=0x03010200:yuva444p
9582 Swap U and V plane in yuv420p stream:
9584 format=yuv420p,mergeplanes=0x000201:yuv420p
9588 Cast a rgb24 clip to yuv444p:
9590 format=rgb24,mergeplanes=0x000102:yuv444p
9596 Estimate and export motion vectors using block matching algorithms.
9597 Motion vectors are stored in frame side data to be used by other filters.
9599 This filter accepts the following options:
9602 Specify the motion estimation method. Accepts one of the following values:
9606 Exhaustive search algorithm.
9608 Three step search algorithm.
9610 Two dimensional logarithmic search algorithm.
9612 New three step search algorithm.
9614 Four step search algorithm.
9616 Diamond search algorithm.
9618 Hexagon-based search algorithm.
9620 Enhanced predictive zonal search algorithm.
9622 Uneven multi-hexagon search algorithm.
9624 Default value is @samp{esa}.
9627 Macroblock size. Default @code{16}.
9630 Search parameter. Default @code{7}.
9633 @section midequalizer
9635 Apply Midway Image Equalization effect using two video streams.
9637 Midway Image Equalization adjusts a pair of images to have the same
9638 histogram, while maintaining their dynamics as much as possible. It's
9639 useful for e.g. matching exposures from a pair of stereo cameras.
9641 This filter has two inputs and one output, which must be of same pixel format, but
9642 may be of different sizes. The output of filter is first input adjusted with
9643 midway histogram of both inputs.
9645 This filter accepts the following option:
9649 Set which planes to process. Default is @code{15}, which is all available planes.
9652 @section minterpolate
9654 Convert the video to specified frame rate using motion interpolation.
9656 This filter accepts the following options:
9659 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}.
9662 Motion interpolation mode. Following values are accepted:
9665 Duplicate previous or next frame for interpolating new ones.
9667 Blend source frames. Interpolated frame is mean of previous and next frames.
9669 Motion compensated interpolation. Following options are effective when this mode is selected:
9673 Motion compensation mode. Following values are accepted:
9676 Overlapped block motion compensation.
9678 Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9680 Default mode is @samp{obmc}.
9683 Motion estimation mode. Following values are accepted:
9686 Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9688 Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9690 Default mode is @samp{bilat}.
9693 The algorithm to be used for motion estimation. Following values are accepted:
9696 Exhaustive search algorithm.
9698 Three step search algorithm.
9700 Two dimensional logarithmic search algorithm.
9702 New three step search algorithm.
9704 Four step search algorithm.
9706 Diamond search algorithm.
9708 Hexagon-based search algorithm.
9710 Enhanced predictive zonal search algorithm.
9712 Uneven multi-hexagon search algorithm.
9714 Default algorithm is @samp{epzs}.
9717 Macroblock size. Default @code{16}.
9720 Motion estimation search parameter. Default @code{32}.
9723 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).
9728 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:
9731 Disable scene change detection.
9733 Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9735 Default method is @samp{fdiff}.
9738 Scene change detection threshold. Default is @code{5.0}.
9743 Drop frames that do not differ greatly from the previous frame in
9744 order to reduce frame rate.
9746 The main use of this filter is for very-low-bitrate encoding
9747 (e.g. streaming over dialup modem), but it could in theory be used for
9748 fixing movies that were inverse-telecined incorrectly.
9750 A description of the accepted options follows.
9754 Set the maximum number of consecutive frames which can be dropped (if
9755 positive), or the minimum interval between dropped frames (if
9756 negative). If the value is 0, the frame is dropped unregarding the
9757 number of previous sequentially dropped frames.
9764 Set the dropping threshold values.
9766 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9767 represent actual pixel value differences, so a threshold of 64
9768 corresponds to 1 unit of difference for each pixel, or the same spread
9769 out differently over the block.
9771 A frame is a candidate for dropping if no 8x8 blocks differ by more
9772 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9773 meaning the whole image) differ by more than a threshold of @option{lo}.
9775 Default value for @option{hi} is 64*12, default value for @option{lo} is
9776 64*5, and default value for @option{frac} is 0.33.
9784 It accepts an integer in input; if non-zero it negates the
9785 alpha component (if available). The default value in input is 0.
9789 Denoise frames using Non-Local Means algorithm.
9791 Each pixel is adjusted by looking for other pixels with similar contexts. This
9792 context similarity is defined by comparing their surrounding patches of size
9793 @option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9796 Note that the research area defines centers for patches, which means some
9797 patches will be made of pixels outside that research area.
9799 The filter accepts the following options.
9803 Set denoising strength.
9809 Same as @option{p} but for chroma planes.
9811 The default value is @var{0} and means automatic.
9817 Same as @option{r} but for chroma planes.
9819 The default value is @var{0} and means automatic.
9824 Deinterlace video using neural network edge directed interpolation.
9826 This filter accepts the following options:
9830 Mandatory option, without binary file filter can not work.
9831 Currently file can be found here:
9832 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9835 Set which frames to deinterlace, by default it is @code{all}.
9836 Can be @code{all} or @code{interlaced}.
9839 Set mode of operation.
9841 Can be one of the following:
9845 Use frame flags, both fields.
9847 Use frame flags, single field.
9851 Use bottom field only.
9853 Use both fields, top first.
9855 Use both fields, bottom first.
9859 Set which planes to process, by default filter process all frames.
9862 Set size of local neighborhood around each pixel, used by the predictor neural
9865 Can be one of the following:
9878 Set the number of neurons in predicctor neural network.
9879 Can be one of the following:
9890 Controls the number of different neural network predictions that are blended
9891 together to compute the final output value. Can be @code{fast}, default or
9895 Set which set of weights to use in the predictor.
9896 Can be one of the following:
9900 weights trained to minimize absolute error
9902 weights trained to minimize squared error
9906 Controls whether or not the prescreener neural network is used to decide
9907 which pixels should be processed by the predictor neural network and which
9908 can be handled by simple cubic interpolation.
9909 The prescreener is trained to know whether cubic interpolation will be
9910 sufficient for a pixel or whether it should be predicted by the predictor nn.
9911 The computational complexity of the prescreener nn is much less than that of
9912 the predictor nn. Since most pixels can be handled by cubic interpolation,
9913 using the prescreener generally results in much faster processing.
9914 The prescreener is pretty accurate, so the difference between using it and not
9915 using it is almost always unnoticeable.
9917 Can be one of the following:
9925 Default is @code{new}.
9928 Set various debugging flags.
9933 Force libavfilter not to use any of the specified pixel formats for the
9934 input to the next filter.
9936 It accepts the following parameters:
9940 A '|'-separated list of pixel format names, such as
9941 apix_fmts=yuv420p|monow|rgb24".
9945 @subsection Examples
9949 Force libavfilter to use a format different from @var{yuv420p} for the
9950 input to the vflip filter:
9952 noformat=pix_fmts=yuv420p,vflip
9956 Convert the input video to any of the formats not contained in the list:
9958 noformat=yuv420p|yuv444p|yuv410p
9964 Add noise on video input frame.
9966 The filter accepts the following options:
9974 Set noise seed for specific pixel component or all pixel components in case
9975 of @var{all_seed}. Default value is @code{123457}.
9977 @item all_strength, alls
9978 @item c0_strength, c0s
9979 @item c1_strength, c1s
9980 @item c2_strength, c2s
9981 @item c3_strength, c3s
9982 Set noise strength for specific pixel component or all pixel components in case
9983 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
9985 @item all_flags, allf
9990 Set pixel component flags or set flags for all components if @var{all_flags}.
9991 Available values for component flags are:
9994 averaged temporal noise (smoother)
9996 mix random noise with a (semi)regular pattern
9998 temporal noise (noise pattern changes between frames)
10000 uniform noise (gaussian otherwise)
10004 @subsection Examples
10006 Add temporal and uniform noise to input video:
10008 noise=alls=20:allf=t+u
10013 Pass the video source unchanged to the output.
10016 Optical Character Recognition
10018 This filter uses Tesseract for optical character recognition.
10020 It accepts the following options:
10024 Set datapath to tesseract data. Default is to use whatever was
10025 set at installation.
10028 Set language, default is "eng".
10031 Set character whitelist.
10034 Set character blacklist.
10037 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10041 Apply a video transform using libopencv.
10043 To enable this filter, install the libopencv library and headers and
10044 configure FFmpeg with @code{--enable-libopencv}.
10046 It accepts the following parameters:
10051 The name of the libopencv filter to apply.
10053 @item filter_params
10054 The parameters to pass to the libopencv filter. If not specified, the default
10055 values are assumed.
10059 Refer to the official libopencv documentation for more precise
10061 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10063 Several libopencv filters are supported; see the following subsections.
10068 Dilate an image by using a specific structuring element.
10069 It corresponds to the libopencv function @code{cvDilate}.
10071 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10073 @var{struct_el} represents a structuring element, and has the syntax:
10074 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10076 @var{cols} and @var{rows} represent the number of columns and rows of
10077 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10078 point, and @var{shape} the shape for the structuring element. @var{shape}
10079 must be "rect", "cross", "ellipse", or "custom".
10081 If the value for @var{shape} is "custom", it must be followed by a
10082 string of the form "=@var{filename}". The file with name
10083 @var{filename} is assumed to represent a binary image, with each
10084 printable character corresponding to a bright pixel. When a custom
10085 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10086 or columns and rows of the read file are assumed instead.
10088 The default value for @var{struct_el} is "3x3+0x0/rect".
10090 @var{nb_iterations} specifies the number of times the transform is
10091 applied to the image, and defaults to 1.
10095 # Use the default values
10098 # Dilate using a structuring element with a 5x5 cross, iterating two times
10099 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10101 # Read the shape from the file diamond.shape, iterating two times.
10102 # The file diamond.shape may contain a pattern of characters like this
10108 # The specified columns and rows are ignored
10109 # but the anchor point coordinates are not
10110 ocv=dilate:0x0+2x2/custom=diamond.shape|2
10115 Erode an image by using a specific structuring element.
10116 It corresponds to the libopencv function @code{cvErode}.
10118 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10119 with the same syntax and semantics as the @ref{dilate} filter.
10123 Smooth the input video.
10125 The filter takes the following parameters:
10126 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10128 @var{type} is the type of smooth filter to apply, and must be one of
10129 the following values: "blur", "blur_no_scale", "median", "gaussian",
10130 or "bilateral". The default value is "gaussian".
10132 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10133 depend on the smooth type. @var{param1} and
10134 @var{param2} accept integer positive values or 0. @var{param3} and
10135 @var{param4} accept floating point values.
10137 The default value for @var{param1} is 3. The default value for the
10138 other parameters is 0.
10140 These parameters correspond to the parameters assigned to the
10141 libopencv function @code{cvSmooth}.
10146 Overlay one video on top of another.
10148 It takes two inputs and has one output. The first input is the "main"
10149 video on which the second input is overlaid.
10151 It accepts the following parameters:
10153 A description of the accepted options follows.
10158 Set the expression for the x and y coordinates of the overlaid video
10159 on the main video. Default value is "0" for both expressions. In case
10160 the expression is invalid, it is set to a huge value (meaning that the
10161 overlay will not be displayed within the output visible area).
10164 The action to take when EOF is encountered on the secondary input; it accepts
10165 one of the following values:
10169 Repeat the last frame (the default).
10173 Pass the main input through.
10177 Set when the expressions for @option{x}, and @option{y} are evaluated.
10179 It accepts the following values:
10182 only evaluate expressions once during the filter initialization or
10183 when a command is processed
10186 evaluate expressions for each incoming frame
10189 Default value is @samp{frame}.
10192 If set to 1, force the output to terminate when the shortest input
10193 terminates. Default value is 0.
10196 Set the format for the output video.
10198 It accepts the following values:
10201 force YUV420 output
10204 force YUV422 output
10207 force YUV444 output
10210 force packed RGB output
10213 force planar RGB output
10216 Default value is @samp{yuv420}.
10218 @item rgb @emph{(deprecated)}
10219 If set to 1, force the filter to accept inputs in the RGB
10220 color space. Default value is 0. This option is deprecated, use
10221 @option{format} instead.
10224 If set to 1, force the filter to draw the last overlay frame over the
10225 main input until the end of the stream. A value of 0 disables this
10226 behavior. Default value is 1.
10229 The @option{x}, and @option{y} expressions can contain the following
10235 The main input width and height.
10239 The overlay input width and height.
10243 The computed values for @var{x} and @var{y}. They are evaluated for
10248 horizontal and vertical chroma subsample values of the output
10249 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10253 the number of input frame, starting from 0
10256 the position in the file of the input frame, NAN if unknown
10259 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10263 Note that the @var{n}, @var{pos}, @var{t} variables are available only
10264 when evaluation is done @emph{per frame}, and will evaluate to NAN
10265 when @option{eval} is set to @samp{init}.
10267 Be aware that frames are taken from each input video in timestamp
10268 order, hence, if their initial timestamps differ, it is a good idea
10269 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10270 have them begin in the same zero timestamp, as the example for
10271 the @var{movie} filter does.
10273 You can chain together more overlays but you should test the
10274 efficiency of such approach.
10276 @subsection Commands
10278 This filter supports the following commands:
10282 Modify the x and y of the overlay input.
10283 The command accepts the same syntax of the corresponding option.
10285 If the specified expression is not valid, it is kept at its current
10289 @subsection Examples
10293 Draw the overlay at 10 pixels from the bottom right corner of the main
10296 overlay=main_w-overlay_w-10:main_h-overlay_h-10
10299 Using named options the example above becomes:
10301 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10305 Insert a transparent PNG logo in the bottom left corner of the input,
10306 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10308 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10312 Insert 2 different transparent PNG logos (second logo on bottom
10313 right corner) using the @command{ffmpeg} tool:
10315 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
10319 Add a transparent color layer on top of the main video; @code{WxH}
10320 must specify the size of the main input to the overlay filter:
10322 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10326 Play an original video and a filtered version (here with the deshake
10327 filter) side by side using the @command{ffplay} tool:
10329 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10332 The above command is the same as:
10334 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10338 Make a sliding overlay appearing from the left to the right top part of the
10339 screen starting since time 2:
10341 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10345 Compose output by putting two input videos side to side:
10347 ffmpeg -i left.avi -i right.avi -filter_complex "
10348 nullsrc=size=200x100 [background];
10349 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10350 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10351 [background][left] overlay=shortest=1 [background+left];
10352 [background+left][right] overlay=shortest=1:x=100 [left+right]
10357 Mask 10-20 seconds of a video by applying the delogo filter to a section
10359 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10360 -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]'
10365 Chain several overlays in cascade:
10367 nullsrc=s=200x200 [bg];
10368 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10369 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10370 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10371 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10372 [in3] null, [mid2] overlay=100:100 [out0]
10379 Apply Overcomplete Wavelet denoiser.
10381 The filter accepts the following options:
10387 Larger depth values will denoise lower frequency components more, but
10388 slow down filtering.
10390 Must be an int in the range 8-16, default is @code{8}.
10392 @item luma_strength, ls
10395 Must be a double value in the range 0-1000, default is @code{1.0}.
10397 @item chroma_strength, cs
10398 Set chroma strength.
10400 Must be a double value in the range 0-1000, default is @code{1.0}.
10406 Add paddings to the input image, and place the original input at the
10407 provided @var{x}, @var{y} coordinates.
10409 It accepts the following parameters:
10414 Specify an expression for the size of the output image with the
10415 paddings added. If the value for @var{width} or @var{height} is 0, the
10416 corresponding input size is used for the output.
10418 The @var{width} expression can reference the value set by the
10419 @var{height} expression, and vice versa.
10421 The default value of @var{width} and @var{height} is 0.
10425 Specify the offsets to place the input image at within the padded area,
10426 with respect to the top/left border of the output image.
10428 The @var{x} expression can reference the value set by the @var{y}
10429 expression, and vice versa.
10431 The default value of @var{x} and @var{y} is 0.
10433 If @var{x} or @var{y} evaluate to a negative number, they'll be changed
10434 so the input image is centered on the padded area.
10437 Specify the color of the padded area. For the syntax of this option,
10438 check the "Color" section in the ffmpeg-utils manual.
10440 The default value of @var{color} is "black".
10443 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10445 It accepts the following values:
10449 Only evaluate expressions once during the filter initialization or when
10450 a command is processed.
10453 Evaluate expressions for each incoming frame.
10457 Default value is @samp{init}.
10460 Pad to aspect instead to a resolution.
10464 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10465 options are expressions containing the following constants:
10470 The input video width and height.
10474 These are the same as @var{in_w} and @var{in_h}.
10478 The output width and height (the size of the padded area), as
10479 specified by the @var{width} and @var{height} expressions.
10483 These are the same as @var{out_w} and @var{out_h}.
10487 The x and y offsets as specified by the @var{x} and @var{y}
10488 expressions, or NAN if not yet specified.
10491 same as @var{iw} / @var{ih}
10494 input sample aspect ratio
10497 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10501 The horizontal and vertical chroma subsample values. For example for the
10502 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10505 @subsection Examples
10509 Add paddings with the color "violet" to the input video. The output video
10510 size is 640x480, and the top-left corner of the input video is placed at
10513 pad=640:480:0:40:violet
10516 The example above is equivalent to the following command:
10518 pad=width=640:height=480:x=0:y=40:color=violet
10522 Pad the input to get an output with dimensions increased by 3/2,
10523 and put the input video at the center of the padded area:
10525 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10529 Pad the input to get a squared output with size equal to the maximum
10530 value between the input width and height, and put the input video at
10531 the center of the padded area:
10533 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10537 Pad the input to get a final w/h ratio of 16:9:
10539 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10543 In case of anamorphic video, in order to set the output display aspect
10544 correctly, it is necessary to use @var{sar} in the expression,
10545 according to the relation:
10547 (ih * X / ih) * sar = output_dar
10548 X = output_dar / sar
10551 Thus the previous example needs to be modified to:
10553 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10557 Double the output size and put the input video in the bottom-right
10558 corner of the output padded area:
10560 pad="2*iw:2*ih:ow-iw:oh-ih"
10564 @anchor{palettegen}
10565 @section palettegen
10567 Generate one palette for a whole video stream.
10569 It accepts the following options:
10573 Set the maximum number of colors to quantize in the palette.
10574 Note: the palette will still contain 256 colors; the unused palette entries
10577 @item reserve_transparent
10578 Create a palette of 255 colors maximum and reserve the last one for
10579 transparency. Reserving the transparency color is useful for GIF optimization.
10580 If not set, the maximum of colors in the palette will be 256. You probably want
10581 to disable this option for a standalone image.
10585 Set statistics mode.
10587 It accepts the following values:
10590 Compute full frame histograms.
10592 Compute histograms only for the part that differs from previous frame. This
10593 might be relevant to give more importance to the moving part of your input if
10594 the background is static.
10596 Compute new histogram for each frame.
10599 Default value is @var{full}.
10602 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10603 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10604 color quantization of the palette. This information is also visible at
10605 @var{info} logging level.
10607 @subsection Examples
10611 Generate a representative palette of a given video using @command{ffmpeg}:
10613 ffmpeg -i input.mkv -vf palettegen palette.png
10617 @section paletteuse
10619 Use a palette to downsample an input video stream.
10621 The filter takes two inputs: one video stream and a palette. The palette must
10622 be a 256 pixels image.
10624 It accepts the following options:
10628 Select dithering mode. Available algorithms are:
10631 Ordered 8x8 bayer dithering (deterministic)
10633 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10634 Note: this dithering is sometimes considered "wrong" and is included as a
10636 @item floyd_steinberg
10637 Floyd and Steingberg dithering (error diffusion)
10639 Frankie Sierra dithering v2 (error diffusion)
10641 Frankie Sierra dithering v2 "Lite" (error diffusion)
10644 Default is @var{sierra2_4a}.
10647 When @var{bayer} dithering is selected, this option defines the scale of the
10648 pattern (how much the crosshatch pattern is visible). A low value means more
10649 visible pattern for less banding, and higher value means less visible pattern
10650 at the cost of more banding.
10652 The option must be an integer value in the range [0,5]. Default is @var{2}.
10655 If set, define the zone to process
10659 Only the changing rectangle will be reprocessed. This is similar to GIF
10660 cropping/offsetting compression mechanism. This option can be useful for speed
10661 if only a part of the image is changing, and has use cases such as limiting the
10662 scope of the error diffusal @option{dither} to the rectangle that bounds the
10663 moving scene (it leads to more deterministic output if the scene doesn't change
10664 much, and as a result less moving noise and better GIF compression).
10667 Default is @var{none}.
10670 Take new palette for each output frame.
10673 @subsection Examples
10677 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10678 using @command{ffmpeg}:
10680 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10684 @section perspective
10686 Correct perspective of video not recorded perpendicular to the screen.
10688 A description of the accepted parameters follows.
10699 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10700 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10701 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10702 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10703 then the corners of the source will be sent to the specified coordinates.
10705 The expressions can use the following variables:
10710 the width and height of video frame.
10714 Output frame count.
10717 @item interpolation
10718 Set interpolation for perspective correction.
10720 It accepts the following values:
10726 Default value is @samp{linear}.
10729 Set interpretation of coordinate options.
10731 It accepts the following values:
10735 Send point in the source specified by the given coordinates to
10736 the corners of the destination.
10738 @item 1, destination
10740 Send the corners of the source to the point in the destination specified
10741 by the given coordinates.
10743 Default value is @samp{source}.
10747 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10749 It accepts the following values:
10752 only evaluate expressions once during the filter initialization or
10753 when a command is processed
10756 evaluate expressions for each incoming frame
10759 Default value is @samp{init}.
10764 Delay interlaced video by one field time so that the field order changes.
10766 The intended use is to fix PAL movies that have been captured with the
10767 opposite field order to the film-to-video transfer.
10769 A description of the accepted parameters follows.
10775 It accepts the following values:
10778 Capture field order top-first, transfer bottom-first.
10779 Filter will delay the bottom field.
10782 Capture field order bottom-first, transfer top-first.
10783 Filter will delay the top field.
10786 Capture and transfer with the same field order. This mode only exists
10787 for the documentation of the other options to refer to, but if you
10788 actually select it, the filter will faithfully do nothing.
10791 Capture field order determined automatically by field flags, transfer
10793 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10794 basis using field flags. If no field information is available,
10795 then this works just like @samp{u}.
10798 Capture unknown or varying, transfer opposite.
10799 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10800 analyzing the images and selecting the alternative that produces best
10801 match between the fields.
10804 Capture top-first, transfer unknown or varying.
10805 Filter selects among @samp{t} and @samp{p} using image analysis.
10808 Capture bottom-first, transfer unknown or varying.
10809 Filter selects among @samp{b} and @samp{p} using image analysis.
10812 Capture determined by field flags, transfer unknown or varying.
10813 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10814 image analysis. If no field information is available, then this works just
10815 like @samp{U}. This is the default mode.
10818 Both capture and transfer unknown or varying.
10819 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10823 @section pixdesctest
10825 Pixel format descriptor test filter, mainly useful for internal
10826 testing. The output video should be equal to the input video.
10830 format=monow, pixdesctest
10833 can be used to test the monowhite pixel format descriptor definition.
10837 Enable the specified chain of postprocessing subfilters using libpostproc. This
10838 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10839 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10840 Each subfilter and some options have a short and a long name that can be used
10841 interchangeably, i.e. dr/dering are the same.
10843 The filters accept the following options:
10847 Set postprocessing subfilters string.
10850 All subfilters share common options to determine their scope:
10854 Honor the quality commands for this subfilter.
10857 Do chrominance filtering, too (default).
10860 Do luminance filtering only (no chrominance).
10863 Do chrominance filtering only (no luminance).
10866 These options can be appended after the subfilter name, separated by a '|'.
10868 Available subfilters are:
10871 @item hb/hdeblock[|difference[|flatness]]
10872 Horizontal deblocking filter
10875 Difference factor where higher values mean more deblocking (default: @code{32}).
10877 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10880 @item vb/vdeblock[|difference[|flatness]]
10881 Vertical deblocking filter
10884 Difference factor where higher values mean more deblocking (default: @code{32}).
10886 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10889 @item ha/hadeblock[|difference[|flatness]]
10890 Accurate horizontal deblocking filter
10893 Difference factor where higher values mean more deblocking (default: @code{32}).
10895 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10898 @item va/vadeblock[|difference[|flatness]]
10899 Accurate vertical deblocking filter
10902 Difference factor where higher values mean more deblocking (default: @code{32}).
10904 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10908 The horizontal and vertical deblocking filters share the difference and
10909 flatness values so you cannot set different horizontal and vertical
10913 @item h1/x1hdeblock
10914 Experimental horizontal deblocking filter
10916 @item v1/x1vdeblock
10917 Experimental vertical deblocking filter
10922 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10925 larger -> stronger filtering
10927 larger -> stronger filtering
10929 larger -> stronger filtering
10932 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10935 Stretch luminance to @code{0-255}.
10938 @item lb/linblenddeint
10939 Linear blend deinterlacing filter that deinterlaces the given block by
10940 filtering all lines with a @code{(1 2 1)} filter.
10942 @item li/linipoldeint
10943 Linear interpolating deinterlacing filter that deinterlaces the given block by
10944 linearly interpolating every second line.
10946 @item ci/cubicipoldeint
10947 Cubic interpolating deinterlacing filter deinterlaces the given block by
10948 cubically interpolating every second line.
10950 @item md/mediandeint
10951 Median deinterlacing filter that deinterlaces the given block by applying a
10952 median filter to every second line.
10954 @item fd/ffmpegdeint
10955 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10956 second line with a @code{(-1 4 2 4 -1)} filter.
10959 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10960 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10962 @item fq/forceQuant[|quantizer]
10963 Overrides the quantizer table from the input with the constant quantizer you
10971 Default pp filter combination (@code{hb|a,vb|a,dr|a})
10974 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10977 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10980 @subsection Examples
10984 Apply horizontal and vertical deblocking, deringing and automatic
10985 brightness/contrast:
10991 Apply default filters without brightness/contrast correction:
10997 Apply default filters and temporal denoiser:
10999 pp=default/tmpnoise|1|2|3
11003 Apply deblocking on luminance only, and switch vertical deblocking on or off
11004 automatically depending on available CPU time:
11011 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11012 similar to spp = 6 with 7 point DCT, where only the center sample is
11015 The filter accepts the following options:
11019 Force a constant quantization parameter. It accepts an integer in range
11020 0 to 63. If not set, the filter will use the QP from the video stream
11024 Set thresholding mode. Available modes are:
11028 Set hard thresholding.
11030 Set soft thresholding (better de-ringing effect, but likely blurrier).
11032 Set medium thresholding (good results, default).
11036 @section premultiply
11037 Apply alpha premultiply effect to input video stream using first plane
11038 of second stream as alpha.
11040 Both streams must have same dimensions and same pixel format.
11043 Apply prewitt operator to input video stream.
11045 The filter accepts the following option:
11049 Set which planes will be processed, unprocessed planes will be copied.
11050 By default value 0xf, all planes will be processed.
11053 Set value which will be multiplied with filtered result.
11056 Set value which will be added to filtered result.
11061 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11062 Ratio) between two input videos.
11064 This filter takes in input two input videos, the first input is
11065 considered the "main" source and is passed unchanged to the
11066 output. The second input is used as a "reference" video for computing
11069 Both video inputs must have the same resolution and pixel format for
11070 this filter to work correctly. Also it assumes that both inputs
11071 have the same number of frames, which are compared one by one.
11073 The obtained average PSNR is printed through the logging system.
11075 The filter stores the accumulated MSE (mean squared error) of each
11076 frame, and at the end of the processing it is averaged across all frames
11077 equally, and the following formula is applied to obtain the PSNR:
11080 PSNR = 10*log10(MAX^2/MSE)
11083 Where MAX is the average of the maximum values of each component of the
11086 The description of the accepted parameters follows.
11089 @item stats_file, f
11090 If specified the filter will use the named file to save the PSNR of
11091 each individual frame. When filename equals "-" the data is sent to
11094 @item stats_version
11095 Specifies which version of the stats file format to use. Details of
11096 each format are written below.
11097 Default value is 1.
11099 @item stats_add_max
11100 Determines whether the max value is output to the stats log.
11101 Default value is 0.
11102 Requires stats_version >= 2. If this is set and stats_version < 2,
11103 the filter will return an error.
11106 The file printed if @var{stats_file} is selected, contains a sequence of
11107 key/value pairs of the form @var{key}:@var{value} for each compared
11110 If a @var{stats_version} greater than 1 is specified, a header line precedes
11111 the list of per-frame-pair stats, with key value pairs following the frame
11112 format with the following parameters:
11115 @item psnr_log_version
11116 The version of the log file format. Will match @var{stats_version}.
11119 A comma separated list of the per-frame-pair parameters included in
11123 A description of each shown per-frame-pair parameter follows:
11127 sequential number of the input frame, starting from 1
11130 Mean Square Error pixel-by-pixel average difference of the compared
11131 frames, averaged over all the image components.
11133 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11134 Mean Square Error pixel-by-pixel average difference of the compared
11135 frames for the component specified by the suffix.
11137 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11138 Peak Signal to Noise ratio of the compared frames for the component
11139 specified by the suffix.
11141 @item max_avg, max_y, max_u, max_v
11142 Maximum allowed value for each channel, and average over all
11148 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11149 [main][ref] psnr="stats_file=stats.log" [out]
11152 On this example the input file being processed is compared with the
11153 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11154 is stored in @file{stats.log}.
11159 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11160 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11163 The pullup filter is designed to take advantage of future context in making
11164 its decisions. This filter is stateless in the sense that it does not lock
11165 onto a pattern to follow, but it instead looks forward to the following
11166 fields in order to identify matches and rebuild progressive frames.
11168 To produce content with an even framerate, insert the fps filter after
11169 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11170 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11172 The filter accepts the following options:
11179 These options set the amount of "junk" to ignore at the left, right, top, and
11180 bottom of the image, respectively. Left and right are in units of 8 pixels,
11181 while top and bottom are in units of 2 lines.
11182 The default is 8 pixels on each side.
11185 Set the strict breaks. Setting this option to 1 will reduce the chances of
11186 filter generating an occasional mismatched frame, but it may also cause an
11187 excessive number of frames to be dropped during high motion sequences.
11188 Conversely, setting it to -1 will make filter match fields more easily.
11189 This may help processing of video where there is slight blurring between
11190 the fields, but may also cause there to be interlaced frames in the output.
11191 Default value is @code{0}.
11194 Set the metric plane to use. It accepts the following values:
11200 Use chroma blue plane.
11203 Use chroma red plane.
11206 This option may be set to use chroma plane instead of the default luma plane
11207 for doing filter's computations. This may improve accuracy on very clean
11208 source material, but more likely will decrease accuracy, especially if there
11209 is chroma noise (rainbow effect) or any grayscale video.
11210 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11211 load and make pullup usable in realtime on slow machines.
11214 For best results (without duplicated frames in the output file) it is
11215 necessary to change the output frame rate. For example, to inverse
11216 telecine NTSC input:
11218 ffmpeg -i input -vf pullup -r 24000/1001 ...
11223 Change video quantization parameters (QP).
11225 The filter accepts the following option:
11229 Set expression for quantization parameter.
11232 The expression is evaluated through the eval API and can contain, among others,
11233 the following constants:
11237 1 if index is not 129, 0 otherwise.
11240 Sequentional index starting from -129 to 128.
11243 @subsection Examples
11247 Some equation like:
11255 Flush video frames from internal cache of frames into a random order.
11256 No frame is discarded.
11257 Inspired by @ref{frei0r} nervous filter.
11261 Set size in number of frames of internal cache, in range from @code{2} to
11262 @code{512}. Default is @code{30}.
11265 Set seed for random number generator, must be an integer included between
11266 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11267 less than @code{0}, the filter will try to use a good random seed on a
11271 @section readeia608
11273 Read closed captioning (EIA-608) information from the top lines of a video frame.
11275 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11276 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11277 with EIA-608 data (starting from 0). A description of each metadata value follows:
11280 @item lavfi.readeia608.X.cc
11281 The two bytes stored as EIA-608 data (printed in hexadecimal).
11283 @item lavfi.readeia608.X.line
11284 The number of the line on which the EIA-608 data was identified and read.
11287 This filter accepts the following options:
11291 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11294 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11297 Set minimal acceptable amplitude change for sync codes detection.
11298 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11301 Set the ratio of width reserved for sync code detection.
11302 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11305 Set the max peaks height difference for sync code detection.
11306 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11309 Set max peaks period difference for sync code detection.
11310 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11313 Set the first two max start code bits differences.
11314 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11317 Set the minimum ratio of bits height compared to 3rd start code bit.
11318 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11321 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11324 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11327 Enable checking the parity bit. In the event of a parity error, the filter will output
11328 @code{0x00} for that character. Default is false.
11331 @subsection Examples
11335 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11337 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
11343 Read vertical interval timecode (VITC) information from the top lines of a
11346 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11347 timecode value, if a valid timecode has been detected. Further metadata key
11348 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11349 timecode data has been found or not.
11351 This filter accepts the following options:
11355 Set the maximum number of lines to scan for VITC data. If the value is set to
11356 @code{-1} the full video frame is scanned. Default is @code{45}.
11359 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11360 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11363 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11364 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11367 @subsection Examples
11371 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11372 draw @code{--:--:--:--} as a placeholder:
11374 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11380 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11382 Destination pixel at position (X, Y) will be picked from source (x, y) position
11383 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11384 value for pixel will be used for destination pixel.
11386 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11387 will have Xmap/Ymap video stream dimensions.
11388 Xmap and Ymap input video streams are 16bit depth, single channel.
11390 @section removegrain
11392 The removegrain filter is a spatial denoiser for progressive video.
11396 Set mode for the first plane.
11399 Set mode for the second plane.
11402 Set mode for the third plane.
11405 Set mode for the fourth plane.
11408 Range of mode is from 0 to 24. Description of each mode follows:
11412 Leave input plane unchanged. Default.
11415 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11418 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11421 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11424 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11425 This is equivalent to a median filter.
11428 Line-sensitive clipping giving the minimal change.
11431 Line-sensitive clipping, intermediate.
11434 Line-sensitive clipping, intermediate.
11437 Line-sensitive clipping, intermediate.
11440 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11443 Replaces the target pixel with the closest neighbour.
11446 [1 2 1] horizontal and vertical kernel blur.
11452 Bob mode, interpolates top field from the line where the neighbours
11453 pixels are the closest.
11456 Bob mode, interpolates bottom field from the line where the neighbours
11457 pixels are the closest.
11460 Bob mode, interpolates top field. Same as 13 but with a more complicated
11461 interpolation formula.
11464 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11465 interpolation formula.
11468 Clips the pixel with the minimum and maximum of respectively the maximum and
11469 minimum of each pair of opposite neighbour pixels.
11472 Line-sensitive clipping using opposite neighbours whose greatest distance from
11473 the current pixel is minimal.
11476 Replaces the pixel with the average of its 8 neighbours.
11479 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11482 Clips pixels using the averages of opposite neighbour.
11485 Same as mode 21 but simpler and faster.
11488 Small edge and halo removal, but reputed useless.
11494 @section removelogo
11496 Suppress a TV station logo, using an image file to determine which
11497 pixels comprise the logo. It works by filling in the pixels that
11498 comprise the logo with neighboring pixels.
11500 The filter accepts the following options:
11504 Set the filter bitmap file, which can be any image format supported by
11505 libavformat. The width and height of the image file must match those of the
11506 video stream being processed.
11509 Pixels in the provided bitmap image with a value of zero are not
11510 considered part of the logo, non-zero pixels are considered part of
11511 the logo. If you use white (255) for the logo and black (0) for the
11512 rest, you will be safe. For making the filter bitmap, it is
11513 recommended to take a screen capture of a black frame with the logo
11514 visible, and then using a threshold filter followed by the erode
11515 filter once or twice.
11517 If needed, little splotches can be fixed manually. Remember that if
11518 logo pixels are not covered, the filter quality will be much
11519 reduced. Marking too many pixels as part of the logo does not hurt as
11520 much, but it will increase the amount of blurring needed to cover over
11521 the image and will destroy more information than necessary, and extra
11522 pixels will slow things down on a large logo.
11524 @section repeatfields
11526 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11527 fields based on its value.
11531 Reverse a video clip.
11533 Warning: This filter requires memory to buffer the entire clip, so trimming
11536 @subsection Examples
11540 Take the first 5 seconds of a clip, and reverse it.
11548 Rotate video by an arbitrary angle expressed in radians.
11550 The filter accepts the following options:
11552 A description of the optional parameters follows.
11555 Set an expression for the angle by which to rotate the input video
11556 clockwise, expressed as a number of radians. A negative value will
11557 result in a counter-clockwise rotation. By default it is set to "0".
11559 This expression is evaluated for each frame.
11562 Set the output width expression, default value is "iw".
11563 This expression is evaluated just once during configuration.
11566 Set the output height expression, default value is "ih".
11567 This expression is evaluated just once during configuration.
11570 Enable bilinear interpolation if set to 1, a value of 0 disables
11571 it. Default value is 1.
11574 Set the color used to fill the output area not covered by the rotated
11575 image. For the general syntax of this option, check the "Color" section in the
11576 ffmpeg-utils manual. If the special value "none" is selected then no
11577 background is printed (useful for example if the background is never shown).
11579 Default value is "black".
11582 The expressions for the angle and the output size can contain the
11583 following constants and functions:
11587 sequential number of the input frame, starting from 0. It is always NAN
11588 before the first frame is filtered.
11591 time in seconds of the input frame, it is set to 0 when the filter is
11592 configured. It is always NAN before the first frame is filtered.
11596 horizontal and vertical chroma subsample values. For example for the
11597 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11601 the input video width and height
11605 the output width and height, that is the size of the padded area as
11606 specified by the @var{width} and @var{height} expressions
11610 the minimal width/height required for completely containing the input
11611 video rotated by @var{a} radians.
11613 These are only available when computing the @option{out_w} and
11614 @option{out_h} expressions.
11617 @subsection Examples
11621 Rotate the input by PI/6 radians clockwise:
11627 Rotate the input by PI/6 radians counter-clockwise:
11633 Rotate the input by 45 degrees clockwise:
11639 Apply a constant rotation with period T, starting from an angle of PI/3:
11641 rotate=PI/3+2*PI*t/T
11645 Make the input video rotation oscillating with a period of T
11646 seconds and an amplitude of A radians:
11648 rotate=A*sin(2*PI/T*t)
11652 Rotate the video, output size is chosen so that the whole rotating
11653 input video is always completely contained in the output:
11655 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11659 Rotate the video, reduce the output size so that no background is ever
11662 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11666 @subsection Commands
11668 The filter supports the following commands:
11672 Set the angle expression.
11673 The command accepts the same syntax of the corresponding option.
11675 If the specified expression is not valid, it is kept at its current
11681 Apply Shape Adaptive Blur.
11683 The filter accepts the following options:
11686 @item luma_radius, lr
11687 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11688 value is 1.0. A greater value will result in a more blurred image, and
11689 in slower processing.
11691 @item luma_pre_filter_radius, lpfr
11692 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11695 @item luma_strength, ls
11696 Set luma maximum difference between pixels to still be considered, must
11697 be a value in the 0.1-100.0 range, default value is 1.0.
11699 @item chroma_radius, cr
11700 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11701 greater value will result in a more blurred image, and in slower
11704 @item chroma_pre_filter_radius, cpfr
11705 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11707 @item chroma_strength, cs
11708 Set chroma maximum difference between pixels to still be considered,
11709 must be a value in the -0.9-100.0 range.
11712 Each chroma option value, if not explicitly specified, is set to the
11713 corresponding luma option value.
11718 Scale (resize) the input video, using the libswscale library.
11720 The scale filter forces the output display aspect ratio to be the same
11721 of the input, by changing the output sample aspect ratio.
11723 If the input image format is different from the format requested by
11724 the next filter, the scale filter will convert the input to the
11727 @subsection Options
11728 The filter accepts the following options, or any of the options
11729 supported by the libswscale scaler.
11731 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11732 the complete list of scaler options.
11737 Set the output video dimension expression. Default value is the input
11740 If the value is 0, the input width is used for the output.
11742 If one of the values is -1, the scale filter will use a value that
11743 maintains the aspect ratio of the input image, calculated from the
11744 other specified dimension. If both of them are -1, the input size is
11747 If one of the values is -n with n > 1, the scale filter will also use a value
11748 that maintains the aspect ratio of the input image, calculated from the other
11749 specified dimension. After that it will, however, make sure that the calculated
11750 dimension is divisible by n and adjust the value if necessary.
11752 See below for the list of accepted constants for use in the dimension
11756 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11760 Only evaluate expressions once during the filter initialization or when a command is processed.
11763 Evaluate expressions for each incoming frame.
11767 Default value is @samp{init}.
11771 Set the interlacing mode. It accepts the following values:
11775 Force interlaced aware scaling.
11778 Do not apply interlaced scaling.
11781 Select interlaced aware scaling depending on whether the source frames
11782 are flagged as interlaced or not.
11785 Default value is @samp{0}.
11788 Set libswscale scaling flags. See
11789 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11790 complete list of values. If not explicitly specified the filter applies
11794 @item param0, param1
11795 Set libswscale input parameters for scaling algorithms that need them. See
11796 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11797 complete documentation. If not explicitly specified the filter applies
11803 Set the video size. For the syntax of this option, check the
11804 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11806 @item in_color_matrix
11807 @item out_color_matrix
11808 Set in/output YCbCr color space type.
11810 This allows the autodetected value to be overridden as well as allows forcing
11811 a specific value used for the output and encoder.
11813 If not specified, the color space type depends on the pixel format.
11819 Choose automatically.
11822 Format conforming to International Telecommunication Union (ITU)
11823 Recommendation BT.709.
11826 Set color space conforming to the United States Federal Communications
11827 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11830 Set color space conforming to:
11834 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11837 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11840 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11845 Set color space conforming to SMPTE ST 240:1999.
11850 Set in/output YCbCr sample range.
11852 This allows the autodetected value to be overridden as well as allows forcing
11853 a specific value used for the output and encoder. If not specified, the
11854 range depends on the pixel format. Possible values:
11858 Choose automatically.
11861 Set full range (0-255 in case of 8-bit luma).
11864 Set "MPEG" range (16-235 in case of 8-bit luma).
11867 @item force_original_aspect_ratio
11868 Enable decreasing or increasing output video width or height if necessary to
11869 keep the original aspect ratio. Possible values:
11873 Scale the video as specified and disable this feature.
11876 The output video dimensions will automatically be decreased if needed.
11879 The output video dimensions will automatically be increased if needed.
11883 One useful instance of this option is that when you know a specific device's
11884 maximum allowed resolution, you can use this to limit the output video to
11885 that, while retaining the aspect ratio. For example, device A allows
11886 1280x720 playback, and your video is 1920x800. Using this option (set it to
11887 decrease) and specifying 1280x720 to the command line makes the output
11890 Please note that this is a different thing than specifying -1 for @option{w}
11891 or @option{h}, you still need to specify the output resolution for this option
11896 The values of the @option{w} and @option{h} options are expressions
11897 containing the following constants:
11902 The input width and height
11906 These are the same as @var{in_w} and @var{in_h}.
11910 The output (scaled) width and height
11914 These are the same as @var{out_w} and @var{out_h}
11917 The same as @var{iw} / @var{ih}
11920 input sample aspect ratio
11923 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11927 horizontal and vertical input chroma subsample values. For example for the
11928 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11932 horizontal and vertical output chroma subsample values. For example for the
11933 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11936 @subsection Examples
11940 Scale the input video to a size of 200x100
11945 This is equivalent to:
11956 Specify a size abbreviation for the output size:
11961 which can also be written as:
11967 Scale the input to 2x:
11969 scale=w=2*iw:h=2*ih
11973 The above is the same as:
11975 scale=2*in_w:2*in_h
11979 Scale the input to 2x with forced interlaced scaling:
11981 scale=2*iw:2*ih:interl=1
11985 Scale the input to half size:
11987 scale=w=iw/2:h=ih/2
11991 Increase the width, and set the height to the same size:
11997 Seek Greek harmony:
12004 Increase the height, and set the width to 3/2 of the height:
12006 scale=w=3/2*oh:h=3/5*ih
12010 Increase the size, making the size a multiple of the chroma
12013 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12017 Increase the width to a maximum of 500 pixels,
12018 keeping the same aspect ratio as the input:
12020 scale=w='min(500\, iw*3/2):h=-1'
12024 @subsection Commands
12026 This filter supports the following commands:
12030 Set the output video dimension expression.
12031 The command accepts the same syntax of the corresponding option.
12033 If the specified expression is not valid, it is kept at its current
12039 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12040 format conversion on CUDA video frames. Setting the output width and height
12041 works in the same way as for the @var{scale} filter.
12043 The following additional options are accepted:
12046 The pixel format of the output CUDA frames. If set to the string "same" (the
12047 default), the input format will be kept. Note that automatic format negotiation
12048 and conversion is not yet supported for hardware frames
12051 The interpolation algorithm used for resizing. One of the following:
12058 @item cubic2p_bspline
12059 2-parameter cubic (B=1, C=0)
12061 @item cubic2p_catmullrom
12062 2-parameter cubic (B=0, C=1/2)
12064 @item cubic2p_b05c03
12065 2-parameter cubic (B=1/2, C=3/10)
12077 Scale (resize) the input video, based on a reference video.
12079 See the scale filter for available options, scale2ref supports the same but
12080 uses the reference video instead of the main input as basis.
12082 @subsection Examples
12086 Scale a subtitle stream to match the main video in size before overlaying
12088 'scale2ref[b][a];[a][b]overlay'
12092 @anchor{selectivecolor}
12093 @section selectivecolor
12095 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12096 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12097 by the "purity" of the color (that is, how saturated it already is).
12099 This filter is similar to the Adobe Photoshop Selective Color tool.
12101 The filter accepts the following options:
12104 @item correction_method
12105 Select color correction method.
12107 Available values are:
12110 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12113 Specified adjustments are relative to the original component value.
12115 Default is @code{absolute}.
12117 Adjustments for red pixels (pixels where the red component is the maximum)
12119 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12121 Adjustments for green pixels (pixels where the green component is the maximum)
12123 Adjustments for cyan pixels (pixels where the red component is the minimum)
12125 Adjustments for blue pixels (pixels where the blue component is the maximum)
12127 Adjustments for magenta pixels (pixels where the green component is the minimum)
12129 Adjustments for white pixels (pixels where all components are greater than 128)
12131 Adjustments for all pixels except pure black and pure white
12133 Adjustments for black pixels (pixels where all components are lesser than 128)
12135 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12138 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12139 4 space separated floating point adjustment values in the [-1,1] range,
12140 respectively to adjust the amount of cyan, magenta, yellow and black for the
12141 pixels of its range.
12143 @subsection Examples
12147 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12148 increase magenta by 27% in blue areas:
12150 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12154 Use a Photoshop selective color preset:
12156 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12160 @anchor{separatefields}
12161 @section separatefields
12163 The @code{separatefields} takes a frame-based video input and splits
12164 each frame into its components fields, producing a new half height clip
12165 with twice the frame rate and twice the frame count.
12167 This filter use field-dominance information in frame to decide which
12168 of each pair of fields to place first in the output.
12169 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12171 @section setdar, setsar
12173 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12176 This is done by changing the specified Sample (aka Pixel) Aspect
12177 Ratio, according to the following equation:
12179 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12182 Keep in mind that the @code{setdar} filter does not modify the pixel
12183 dimensions of the video frame. Also, the display aspect ratio set by
12184 this filter may be changed by later filters in the filterchain,
12185 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12188 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12189 the filter output video.
12191 Note that as a consequence of the application of this filter, the
12192 output display aspect ratio will change according to the equation
12195 Keep in mind that the sample aspect ratio set by the @code{setsar}
12196 filter may be changed by later filters in the filterchain, e.g. if
12197 another "setsar" or a "setdar" filter is applied.
12199 It accepts the following parameters:
12202 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12203 Set the aspect ratio used by the filter.
12205 The parameter can be a floating point number string, an expression, or
12206 a string of the form @var{num}:@var{den}, where @var{num} and
12207 @var{den} are the numerator and denominator of the aspect ratio. If
12208 the parameter is not specified, it is assumed the value "0".
12209 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12213 Set the maximum integer value to use for expressing numerator and
12214 denominator when reducing the expressed aspect ratio to a rational.
12215 Default value is @code{100}.
12219 The parameter @var{sar} is an expression containing
12220 the following constants:
12224 These are approximated values for the mathematical constants e
12225 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12228 The input width and height.
12231 These are the same as @var{w} / @var{h}.
12234 The input sample aspect ratio.
12237 The input display aspect ratio. It is the same as
12238 (@var{w} / @var{h}) * @var{sar}.
12241 Horizontal and vertical chroma subsample values. For example, for the
12242 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12245 @subsection Examples
12250 To change the display aspect ratio to 16:9, specify one of the following:
12257 To change the sample aspect ratio to 10:11, specify:
12263 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12264 1000 in the aspect ratio reduction, use the command:
12266 setdar=ratio=16/9:max=1000
12274 Force field for the output video frame.
12276 The @code{setfield} filter marks the interlace type field for the
12277 output frames. It does not change the input frame, but only sets the
12278 corresponding property, which affects how the frame is treated by
12279 following filters (e.g. @code{fieldorder} or @code{yadif}).
12281 The filter accepts the following options:
12286 Available values are:
12290 Keep the same field property.
12293 Mark the frame as bottom-field-first.
12296 Mark the frame as top-field-first.
12299 Mark the frame as progressive.
12305 Show a line containing various information for each input video frame.
12306 The input video is not modified.
12308 The shown line contains a sequence of key/value pairs of the form
12309 @var{key}:@var{value}.
12311 The following values are shown in the output:
12315 The (sequential) number of the input frame, starting from 0.
12318 The Presentation TimeStamp of the input frame, expressed as a number of
12319 time base units. The time base unit depends on the filter input pad.
12322 The Presentation TimeStamp of the input frame, expressed as a number of
12326 The position of the frame in the input stream, or -1 if this information is
12327 unavailable and/or meaningless (for example in case of synthetic video).
12330 The pixel format name.
12333 The sample aspect ratio of the input frame, expressed in the form
12334 @var{num}/@var{den}.
12337 The size of the input frame. For the syntax of this option, check the
12338 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12341 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12342 for bottom field first).
12345 This is 1 if the frame is a key frame, 0 otherwise.
12348 The picture type of the input frame ("I" for an I-frame, "P" for a
12349 P-frame, "B" for a B-frame, or "?" for an unknown type).
12350 Also refer to the documentation of the @code{AVPictureType} enum and of
12351 the @code{av_get_picture_type_char} function defined in
12352 @file{libavutil/avutil.h}.
12355 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12357 @item plane_checksum
12358 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12359 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12362 @section showpalette
12364 Displays the 256 colors palette of each frame. This filter is only relevant for
12365 @var{pal8} pixel format frames.
12367 It accepts the following option:
12371 Set the size of the box used to represent one palette color entry. Default is
12372 @code{30} (for a @code{30x30} pixel box).
12375 @section shuffleframes
12377 Reorder and/or duplicate and/or drop video frames.
12379 It accepts the following parameters:
12383 Set the destination indexes of input frames.
12384 This is space or '|' separated list of indexes that maps input frames to output
12385 frames. Number of indexes also sets maximal value that each index may have.
12386 '-1' index have special meaning and that is to drop frame.
12389 The first frame has the index 0. The default is to keep the input unchanged.
12391 @subsection Examples
12395 Swap second and third frame of every three frames of the input:
12397 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12401 Swap 10th and 1st frame of every ten frames of the input:
12403 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12407 @section shuffleplanes
12409 Reorder and/or duplicate video planes.
12411 It accepts the following parameters:
12416 The index of the input plane to be used as the first output plane.
12419 The index of the input plane to be used as the second output plane.
12422 The index of the input plane to be used as the third output plane.
12425 The index of the input plane to be used as the fourth output plane.
12429 The first plane has the index 0. The default is to keep the input unchanged.
12431 @subsection Examples
12435 Swap the second and third planes of the input:
12437 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12441 @anchor{signalstats}
12442 @section signalstats
12443 Evaluate various visual metrics that assist in determining issues associated
12444 with the digitization of analog video media.
12446 By default the filter will log these metadata values:
12450 Display the minimal Y value contained within the input frame. Expressed in
12454 Display the Y value at the 10% percentile within the input frame. Expressed in
12458 Display the average Y value within the input frame. Expressed in range of
12462 Display the Y value at the 90% percentile within the input frame. Expressed in
12466 Display the maximum Y value contained within the input frame. Expressed in
12470 Display the minimal U value contained within the input frame. Expressed in
12474 Display the U value at the 10% percentile within the input frame. Expressed in
12478 Display the average U value within the input frame. Expressed in range of
12482 Display the U value at the 90% percentile within the input frame. Expressed in
12486 Display the maximum U value contained within the input frame. Expressed in
12490 Display the minimal V value contained within the input frame. Expressed in
12494 Display the V value at the 10% percentile within the input frame. Expressed in
12498 Display the average V value within the input frame. Expressed in range of
12502 Display the V value at the 90% percentile within the input frame. Expressed in
12506 Display the maximum V value contained within the input frame. Expressed in
12510 Display the minimal saturation value contained within the input frame.
12511 Expressed in range of [0-~181.02].
12514 Display the saturation value at the 10% percentile within the input frame.
12515 Expressed in range of [0-~181.02].
12518 Display the average saturation value within the input frame. Expressed in range
12522 Display the saturation value at the 90% percentile within the input frame.
12523 Expressed in range of [0-~181.02].
12526 Display the maximum saturation value contained within the input frame.
12527 Expressed in range of [0-~181.02].
12530 Display the median value for hue within the input frame. Expressed in range of
12534 Display the average value for hue within the input frame. Expressed in range of
12538 Display the average of sample value difference between all values of the Y
12539 plane in the current frame and corresponding values of the previous input frame.
12540 Expressed in range of [0-255].
12543 Display the average of sample value difference between all values of the U
12544 plane in the current frame and corresponding values of the previous input frame.
12545 Expressed in range of [0-255].
12548 Display the average of sample value difference between all values of the V
12549 plane in the current frame and corresponding values of the previous input frame.
12550 Expressed in range of [0-255].
12553 Display bit depth of Y plane in current frame.
12554 Expressed in range of [0-16].
12557 Display bit depth of U plane in current frame.
12558 Expressed in range of [0-16].
12561 Display bit depth of V plane in current frame.
12562 Expressed in range of [0-16].
12565 The filter accepts the following options:
12571 @option{stat} specify an additional form of image analysis.
12572 @option{out} output video with the specified type of pixel highlighted.
12574 Both options accept the following values:
12578 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12579 unlike the neighboring pixels of the same field. Examples of temporal outliers
12580 include the results of video dropouts, head clogs, or tape tracking issues.
12583 Identify @var{vertical line repetition}. Vertical line repetition includes
12584 similar rows of pixels within a frame. In born-digital video vertical line
12585 repetition is common, but this pattern is uncommon in video digitized from an
12586 analog source. When it occurs in video that results from the digitization of an
12587 analog source it can indicate concealment from a dropout compensator.
12590 Identify pixels that fall outside of legal broadcast range.
12594 Set the highlight color for the @option{out} option. The default color is
12598 @subsection Examples
12602 Output data of various video metrics:
12604 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12608 Output specific data about the minimum and maximum values of the Y plane per frame:
12610 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12614 Playback video while highlighting pixels that are outside of broadcast range in red.
12616 ffplay example.mov -vf signalstats="out=brng:color=red"
12620 Playback video with signalstats metadata drawn over the frame.
12622 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12625 The contents of signalstat_drawtext.txt used in the command are:
12628 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12629 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12630 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12631 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12639 Calculates the MPEG-7 Video Signature. The filter can handle more than one
12640 input. In this case the matching between the inputs can be calculated additionally.
12641 The filter always passes through the first input. The signature of each stream can
12642 be written into a file.
12644 It accepts the following options:
12648 Enable or disable the matching process.
12650 Available values are:
12654 Disable the calculation of a matching (default).
12656 Calculate the matching for the whole video and output whether the whole video
12657 matches or only parts.
12659 Calculate only until a matching is found or the video ends. Should be faster in
12664 Set the number of inputs. The option value must be a non negative integer.
12665 Default value is 1.
12668 Set the path to which the output is written. If there is more than one input,
12669 the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12670 integer), that will be replaced with the input number. If no filename is
12671 specified, no output will be written. This is the default.
12674 Choose the output format.
12676 Available values are:
12680 Use the specified binary representation (default).
12682 Use the specified xml representation.
12686 Set threshold to detect one word as similar. The option value must be an integer
12687 greater than zero. The default value is 9000.
12690 Set threshold to detect all words as similar. The option value must be an integer
12691 greater than zero. The default value is 60000.
12694 Set threshold to detect frames as similar. The option value must be an integer
12695 greater than zero. The default value is 116.
12698 Set the minimum length of a sequence in frames to recognize it as matching
12699 sequence. The option value must be a non negative integer value.
12700 The default value is 0.
12703 Set the minimum relation, that matching frames to all frames must have.
12704 The option value must be a double value between 0 and 1. The default value is 0.5.
12707 @subsection Examples
12711 To calculate the signature of an input video and store it in signature.bin:
12713 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12717 To detect whether two videos match and store the signatures in XML format in
12718 signature0.xml and signature1.xml:
12720 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 -
12728 Blur the input video without impacting the outlines.
12730 It accepts the following options:
12733 @item luma_radius, lr
12734 Set the luma radius. The option value must be a float number in
12735 the range [0.1,5.0] that specifies the variance of the gaussian filter
12736 used to blur the image (slower if larger). Default value is 1.0.
12738 @item luma_strength, ls
12739 Set the luma strength. The option value must be a float number
12740 in the range [-1.0,1.0] that configures the blurring. A value included
12741 in [0.0,1.0] will blur the image whereas a value included in
12742 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12744 @item luma_threshold, lt
12745 Set the luma threshold used as a coefficient to determine
12746 whether a pixel should be blurred or not. The option value must be an
12747 integer in the range [-30,30]. A value of 0 will filter all the image,
12748 a value included in [0,30] will filter flat areas and a value included
12749 in [-30,0] will filter edges. Default value is 0.
12751 @item chroma_radius, cr
12752 Set the chroma radius. The option value must be a float number in
12753 the range [0.1,5.0] that specifies the variance of the gaussian filter
12754 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12756 @item chroma_strength, cs
12757 Set the chroma strength. The option value must be a float number
12758 in the range [-1.0,1.0] that configures the blurring. A value included
12759 in [0.0,1.0] will blur the image whereas a value included in
12760 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12762 @item chroma_threshold, ct
12763 Set the chroma threshold used as a coefficient to determine
12764 whether a pixel should be blurred or not. The option value must be an
12765 integer in the range [-30,30]. A value of 0 will filter all the image,
12766 a value included in [0,30] will filter flat areas and a value included
12767 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12770 If a chroma option is not explicitly set, the corresponding luma value
12775 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12777 This filter takes in input two input videos, the first input is
12778 considered the "main" source and is passed unchanged to the
12779 output. The second input is used as a "reference" video for computing
12782 Both video inputs must have the same resolution and pixel format for
12783 this filter to work correctly. Also it assumes that both inputs
12784 have the same number of frames, which are compared one by one.
12786 The filter stores the calculated SSIM of each frame.
12788 The description of the accepted parameters follows.
12791 @item stats_file, f
12792 If specified the filter will use the named file to save the SSIM of
12793 each individual frame. When filename equals "-" the data is sent to
12797 The file printed if @var{stats_file} is selected, contains a sequence of
12798 key/value pairs of the form @var{key}:@var{value} for each compared
12801 A description of each shown parameter follows:
12805 sequential number of the input frame, starting from 1
12807 @item Y, U, V, R, G, B
12808 SSIM of the compared frames for the component specified by the suffix.
12811 SSIM of the compared frames for the whole frame.
12814 Same as above but in dB representation.
12819 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12820 [main][ref] ssim="stats_file=stats.log" [out]
12823 On this example the input file being processed is compared with the
12824 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12825 is stored in @file{stats.log}.
12827 Another example with both psnr and ssim at same time:
12829 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12834 Convert between different stereoscopic image formats.
12836 The filters accept the following options:
12840 Set stereoscopic image format of input.
12842 Available values for input image formats are:
12845 side by side parallel (left eye left, right eye right)
12848 side by side crosseye (right eye left, left eye right)
12851 side by side parallel with half width resolution
12852 (left eye left, right eye right)
12855 side by side crosseye with half width resolution
12856 (right eye left, left eye right)
12859 above-below (left eye above, right eye below)
12862 above-below (right eye above, left eye below)
12865 above-below with half height resolution
12866 (left eye above, right eye below)
12869 above-below with half height resolution
12870 (right eye above, left eye below)
12873 alternating frames (left eye first, right eye second)
12876 alternating frames (right eye first, left eye second)
12879 interleaved rows (left eye has top row, right eye starts on next row)
12882 interleaved rows (right eye has top row, left eye starts on next row)
12885 interleaved columns, left eye first
12888 interleaved columns, right eye first
12890 Default value is @samp{sbsl}.
12894 Set stereoscopic image format of output.
12898 side by side parallel (left eye left, right eye right)
12901 side by side crosseye (right eye left, left eye right)
12904 side by side parallel with half width resolution
12905 (left eye left, right eye right)
12908 side by side crosseye with half width resolution
12909 (right eye left, left eye right)
12912 above-below (left eye above, right eye below)
12915 above-below (right eye above, left eye below)
12918 above-below with half height resolution
12919 (left eye above, right eye below)
12922 above-below with half height resolution
12923 (right eye above, left eye below)
12926 alternating frames (left eye first, right eye second)
12929 alternating frames (right eye first, left eye second)
12932 interleaved rows (left eye has top row, right eye starts on next row)
12935 interleaved rows (right eye has top row, left eye starts on next row)
12938 anaglyph red/blue gray
12939 (red filter on left eye, blue filter on right eye)
12942 anaglyph red/green gray
12943 (red filter on left eye, green filter on right eye)
12946 anaglyph red/cyan gray
12947 (red filter on left eye, cyan filter on right eye)
12950 anaglyph red/cyan half colored
12951 (red filter on left eye, cyan filter on right eye)
12954 anaglyph red/cyan color
12955 (red filter on left eye, cyan filter on right eye)
12958 anaglyph red/cyan color optimized with the least squares projection of dubois
12959 (red filter on left eye, cyan filter on right eye)
12962 anaglyph green/magenta gray
12963 (green filter on left eye, magenta filter on right eye)
12966 anaglyph green/magenta half colored
12967 (green filter on left eye, magenta filter on right eye)
12970 anaglyph green/magenta colored
12971 (green filter on left eye, magenta filter on right eye)
12974 anaglyph green/magenta color optimized with the least squares projection of dubois
12975 (green filter on left eye, magenta filter on right eye)
12978 anaglyph yellow/blue gray
12979 (yellow filter on left eye, blue filter on right eye)
12982 anaglyph yellow/blue half colored
12983 (yellow filter on left eye, blue filter on right eye)
12986 anaglyph yellow/blue colored
12987 (yellow filter on left eye, blue filter on right eye)
12990 anaglyph yellow/blue color optimized with the least squares projection of dubois
12991 (yellow filter on left eye, blue filter on right eye)
12994 mono output (left eye only)
12997 mono output (right eye only)
13000 checkerboard, left eye first
13003 checkerboard, right eye first
13006 interleaved columns, left eye first
13009 interleaved columns, right eye first
13015 Default value is @samp{arcd}.
13018 @subsection Examples
13022 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13028 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13034 @section streamselect, astreamselect
13035 Select video or audio streams.
13037 The filter accepts the following options:
13041 Set number of inputs. Default is 2.
13044 Set input indexes to remap to outputs.
13047 @subsection Commands
13049 The @code{streamselect} and @code{astreamselect} filter supports the following
13054 Set input indexes to remap to outputs.
13057 @subsection Examples
13061 Select first 5 seconds 1st stream and rest of time 2nd stream:
13063 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13067 Same as above, but for audio:
13069 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13074 Apply sobel operator to input video stream.
13076 The filter accepts the following option:
13080 Set which planes will be processed, unprocessed planes will be copied.
13081 By default value 0xf, all planes will be processed.
13084 Set value which will be multiplied with filtered result.
13087 Set value which will be added to filtered result.
13093 Apply a simple postprocessing filter that compresses and decompresses the image
13094 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13095 and average the results.
13097 The filter accepts the following options:
13101 Set quality. This option defines the number of levels for averaging. It accepts
13102 an integer in the range 0-6. If set to @code{0}, the filter will have no
13103 effect. A value of @code{6} means the higher quality. For each increment of
13104 that value the speed drops by a factor of approximately 2. Default value is
13108 Force a constant quantization parameter. If not set, the filter will use the QP
13109 from the video stream (if available).
13112 Set thresholding mode. Available modes are:
13116 Set hard thresholding (default).
13118 Set soft thresholding (better de-ringing effect, but likely blurrier).
13121 @item use_bframe_qp
13122 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13123 option may cause flicker since the B-Frames have often larger QP. Default is
13124 @code{0} (not enabled).
13130 Draw subtitles on top of input video using the libass library.
13132 To enable compilation of this filter you need to configure FFmpeg with
13133 @code{--enable-libass}. This filter also requires a build with libavcodec and
13134 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13135 Alpha) subtitles format.
13137 The filter accepts the following options:
13141 Set the filename of the subtitle file to read. It must be specified.
13143 @item original_size
13144 Specify the size of the original video, the video for which the ASS file
13145 was composed. For the syntax of this option, check the
13146 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13147 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13148 correctly scale the fonts if the aspect ratio has been changed.
13151 Set a directory path containing fonts that can be used by the filter.
13152 These fonts will be used in addition to whatever the font provider uses.
13155 Set subtitles input character encoding. @code{subtitles} filter only. Only
13156 useful if not UTF-8.
13158 @item stream_index, si
13159 Set subtitles stream index. @code{subtitles} filter only.
13162 Override default style or script info parameters of the subtitles. It accepts a
13163 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13166 If the first key is not specified, it is assumed that the first value
13167 specifies the @option{filename}.
13169 For example, to render the file @file{sub.srt} on top of the input
13170 video, use the command:
13175 which is equivalent to:
13177 subtitles=filename=sub.srt
13180 To render the default subtitles stream from file @file{video.mkv}, use:
13182 subtitles=video.mkv
13185 To render the second subtitles stream from that file, use:
13187 subtitles=video.mkv:si=1
13190 To make the subtitles stream from @file{sub.srt} appear in transparent green
13191 @code{DejaVu Serif}, use:
13193 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13196 @section super2xsai
13198 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13199 Interpolate) pixel art scaling algorithm.
13201 Useful for enlarging pixel art images without reducing sharpness.
13205 Swap two rectangular objects in video.
13207 This filter accepts the following options:
13217 Set 1st rect x coordinate.
13220 Set 1st rect y coordinate.
13223 Set 2nd rect x coordinate.
13226 Set 2nd rect y coordinate.
13228 All expressions are evaluated once for each frame.
13231 The all options are expressions containing the following constants:
13236 The input width and height.
13239 same as @var{w} / @var{h}
13242 input sample aspect ratio
13245 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13248 The number of the input frame, starting from 0.
13251 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13254 the position in the file of the input frame, NAN if unknown
13262 Apply telecine process to the video.
13264 This filter accepts the following options:
13273 The default value is @code{top}.
13277 A string of numbers representing the pulldown pattern you wish to apply.
13278 The default value is @code{23}.
13282 Some typical patterns:
13287 24p: 2332 (preferred)
13294 24p: 222222222223 ("Euro pulldown")
13301 Apply threshold effect to video stream.
13303 This filter needs four video streams to perform thresholding.
13304 First stream is stream we are filtering.
13305 Second stream is holding threshold values, third stream is holding min values,
13306 and last, fourth stream is holding max values.
13308 The filter accepts the following option:
13312 Set which planes will be processed, unprocessed planes will be copied.
13313 By default value 0xf, all planes will be processed.
13316 For example if first stream pixel's component value is less then threshold value
13317 of pixel component from 2nd threshold stream, third stream value will picked,
13318 otherwise fourth stream pixel component value will be picked.
13320 Using color source filter one can perform various types of thresholding:
13322 @subsection Examples
13326 Binary threshold, using gray color as threshold:
13328 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13332 Inverted binary threshold, using gray color as threshold:
13334 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13338 Truncate binary threshold, using gray color as threshold:
13340 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13344 Threshold to zero, using gray color as threshold:
13346 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13350 Inverted threshold to zero, using gray color as threshold:
13352 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13357 Select the most representative frame in a given sequence of consecutive frames.
13359 The filter accepts the following options:
13363 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13364 will pick one of them, and then handle the next batch of @var{n} frames until
13365 the end. Default is @code{100}.
13368 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13369 value will result in a higher memory usage, so a high value is not recommended.
13371 @subsection Examples
13375 Extract one picture each 50 frames:
13381 Complete example of a thumbnail creation with @command{ffmpeg}:
13383 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13389 Tile several successive frames together.
13391 The filter accepts the following options:
13396 Set the grid size (i.e. the number of lines and columns). For the syntax of
13397 this option, check the
13398 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13401 Set the maximum number of frames to render in the given area. It must be less
13402 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13403 the area will be used.
13406 Set the outer border margin in pixels.
13409 Set the inner border thickness (i.e. the number of pixels between frames). For
13410 more advanced padding options (such as having different values for the edges),
13411 refer to the pad video filter.
13414 Specify the color of the unused area. For the syntax of this option, check the
13415 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13419 @subsection Examples
13423 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13425 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13427 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13428 duplicating each output frame to accommodate the originally detected frame
13432 Display @code{5} pictures in an area of @code{3x2} frames,
13433 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13434 mixed flat and named options:
13436 tile=3x2:nb_frames=5:padding=7:margin=2
13440 @section tinterlace
13442 Perform various types of temporal field interlacing.
13444 Frames are counted starting from 1, so the first input frame is
13447 The filter accepts the following options:
13452 Specify the mode of the interlacing. This option can also be specified
13453 as a value alone. See below for a list of values for this option.
13455 Available values are:
13459 Move odd frames into the upper field, even into the lower field,
13460 generating a double height frame at half frame rate.
13464 Frame 1 Frame 2 Frame 3 Frame 4
13466 11111 22222 33333 44444
13467 11111 22222 33333 44444
13468 11111 22222 33333 44444
13469 11111 22222 33333 44444
13483 Only output odd frames, even frames are dropped, generating a frame with
13484 unchanged height at half frame rate.
13489 Frame 1 Frame 2 Frame 3 Frame 4
13491 11111 22222 33333 44444
13492 11111 22222 33333 44444
13493 11111 22222 33333 44444
13494 11111 22222 33333 44444
13504 Only output even frames, odd frames are dropped, generating a frame with
13505 unchanged height at half frame rate.
13510 Frame 1 Frame 2 Frame 3 Frame 4
13512 11111 22222 33333 44444
13513 11111 22222 33333 44444
13514 11111 22222 33333 44444
13515 11111 22222 33333 44444
13525 Expand each frame to full height, but pad alternate lines with black,
13526 generating a frame with double height at the same input frame rate.
13531 Frame 1 Frame 2 Frame 3 Frame 4
13533 11111 22222 33333 44444
13534 11111 22222 33333 44444
13535 11111 22222 33333 44444
13536 11111 22222 33333 44444
13539 11111 ..... 33333 .....
13540 ..... 22222 ..... 44444
13541 11111 ..... 33333 .....
13542 ..... 22222 ..... 44444
13543 11111 ..... 33333 .....
13544 ..... 22222 ..... 44444
13545 11111 ..... 33333 .....
13546 ..... 22222 ..... 44444
13550 @item interleave_top, 4
13551 Interleave the upper field from odd frames with the lower field from
13552 even frames, generating a frame with unchanged height at half frame rate.
13557 Frame 1 Frame 2 Frame 3 Frame 4
13559 11111<- 22222 33333<- 44444
13560 11111 22222<- 33333 44444<-
13561 11111<- 22222 33333<- 44444
13562 11111 22222<- 33333 44444<-
13572 @item interleave_bottom, 5
13573 Interleave the lower field from odd frames with the upper field from
13574 even frames, generating a frame with unchanged height at half frame rate.
13579 Frame 1 Frame 2 Frame 3 Frame 4
13581 11111 22222<- 33333 44444<-
13582 11111<- 22222 33333<- 44444
13583 11111 22222<- 33333 44444<-
13584 11111<- 22222 33333<- 44444
13594 @item interlacex2, 6
13595 Double frame rate with unchanged height. Frames are inserted each
13596 containing the second temporal field from the previous input frame and
13597 the first temporal field from the next input frame. This mode relies on
13598 the top_field_first flag. Useful for interlaced video displays with no
13599 field synchronisation.
13604 Frame 1 Frame 2 Frame 3 Frame 4
13606 11111 22222 33333 44444
13607 11111 22222 33333 44444
13608 11111 22222 33333 44444
13609 11111 22222 33333 44444
13612 11111 22222 22222 33333 33333 44444 44444
13613 11111 11111 22222 22222 33333 33333 44444
13614 11111 22222 22222 33333 33333 44444 44444
13615 11111 11111 22222 22222 33333 33333 44444
13620 Move odd frames into the upper field, even into the lower field,
13621 generating a double height frame at same frame rate.
13626 Frame 1 Frame 2 Frame 3 Frame 4
13628 11111 22222 33333 44444
13629 11111 22222 33333 44444
13630 11111 22222 33333 44444
13631 11111 22222 33333 44444
13634 11111 33333 33333 55555
13635 22222 22222 44444 44444
13636 11111 33333 33333 55555
13637 22222 22222 44444 44444
13638 11111 33333 33333 55555
13639 22222 22222 44444 44444
13640 11111 33333 33333 55555
13641 22222 22222 44444 44444
13646 Numeric values are deprecated but are accepted for backward
13647 compatibility reasons.
13649 Default mode is @code{merge}.
13652 Specify flags influencing the filter process.
13654 Available value for @var{flags} is:
13657 @item low_pass_filter, vlfp
13658 Enable vertical low-pass filtering in the filter.
13659 Vertical low-pass filtering is required when creating an interlaced
13660 destination from a progressive source which contains high-frequency
13661 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13664 Vertical low-pass filtering can only be enabled for @option{mode}
13665 @var{interleave_top} and @var{interleave_bottom}.
13672 Transpose rows with columns in the input video and optionally flip it.
13674 It accepts the following parameters:
13679 Specify the transposition direction.
13681 Can assume the following values:
13683 @item 0, 4, cclock_flip
13684 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13692 Rotate by 90 degrees clockwise, that is:
13700 Rotate by 90 degrees counterclockwise, that is:
13707 @item 3, 7, clock_flip
13708 Rotate by 90 degrees clockwise and vertically flip, that is:
13716 For values between 4-7, the transposition is only done if the input
13717 video geometry is portrait and not landscape. These values are
13718 deprecated, the @code{passthrough} option should be used instead.
13720 Numerical values are deprecated, and should be dropped in favor of
13721 symbolic constants.
13724 Do not apply the transposition if the input geometry matches the one
13725 specified by the specified value. It accepts the following values:
13728 Always apply transposition.
13730 Preserve portrait geometry (when @var{height} >= @var{width}).
13732 Preserve landscape geometry (when @var{width} >= @var{height}).
13735 Default value is @code{none}.
13738 For example to rotate by 90 degrees clockwise and preserve portrait
13741 transpose=dir=1:passthrough=portrait
13744 The command above can also be specified as:
13746 transpose=1:portrait
13750 Trim the input so that the output contains one continuous subpart of the input.
13752 It accepts the following parameters:
13755 Specify the time of the start of the kept section, i.e. the frame with the
13756 timestamp @var{start} will be the first frame in the output.
13759 Specify the time of the first frame that will be dropped, i.e. the frame
13760 immediately preceding the one with the timestamp @var{end} will be the last
13761 frame in the output.
13764 This is the same as @var{start}, except this option sets the start timestamp
13765 in timebase units instead of seconds.
13768 This is the same as @var{end}, except this option sets the end timestamp
13769 in timebase units instead of seconds.
13772 The maximum duration of the output in seconds.
13775 The number of the first frame that should be passed to the output.
13778 The number of the first frame that should be dropped.
13781 @option{start}, @option{end}, and @option{duration} are expressed as time
13782 duration specifications; see
13783 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13784 for the accepted syntax.
13786 Note that the first two sets of the start/end options and the @option{duration}
13787 option look at the frame timestamp, while the _frame variants simply count the
13788 frames that pass through the filter. Also note that this filter does not modify
13789 the timestamps. If you wish for the output timestamps to start at zero, insert a
13790 setpts filter after the trim filter.
13792 If multiple start or end options are set, this filter tries to be greedy and
13793 keep all the frames that match at least one of the specified constraints. To keep
13794 only the part that matches all the constraints at once, chain multiple trim
13797 The defaults are such that all the input is kept. So it is possible to set e.g.
13798 just the end values to keep everything before the specified time.
13803 Drop everything except the second minute of input:
13805 ffmpeg -i INPUT -vf trim=60:120
13809 Keep only the first second:
13811 ffmpeg -i INPUT -vf trim=duration=1
13820 Sharpen or blur the input video.
13822 It accepts the following parameters:
13825 @item luma_msize_x, lx
13826 Set the luma matrix horizontal size. It must be an odd integer between
13827 3 and 23. The default value is 5.
13829 @item luma_msize_y, ly
13830 Set the luma matrix vertical size. It must be an odd integer between 3
13831 and 23. The default value is 5.
13833 @item luma_amount, la
13834 Set the luma effect strength. It must be a floating point number, reasonable
13835 values lay between -1.5 and 1.5.
13837 Negative values will blur the input video, while positive values will
13838 sharpen it, a value of zero will disable the effect.
13840 Default value is 1.0.
13842 @item chroma_msize_x, cx
13843 Set the chroma matrix horizontal size. It must be an odd integer
13844 between 3 and 23. The default value is 5.
13846 @item chroma_msize_y, cy
13847 Set the chroma matrix vertical size. It must be an odd integer
13848 between 3 and 23. The default value is 5.
13850 @item chroma_amount, ca
13851 Set the chroma effect strength. It must be a floating point number, reasonable
13852 values lay between -1.5 and 1.5.
13854 Negative values will blur the input video, while positive values will
13855 sharpen it, a value of zero will disable the effect.
13857 Default value is 0.0.
13860 If set to 1, specify using OpenCL capabilities, only available if
13861 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13865 All parameters are optional and default to the equivalent of the
13866 string '5:5:1.0:5:5:0.0'.
13868 @subsection Examples
13872 Apply strong luma sharpen effect:
13874 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13878 Apply a strong blur of both luma and chroma parameters:
13880 unsharp=7:7:-2:7:7:-2
13886 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13887 the image at several (or - in the case of @option{quality} level @code{8} - all)
13888 shifts and average the results.
13890 The way this differs from the behavior of spp is that uspp actually encodes &
13891 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13892 DCT similar to MJPEG.
13894 The filter accepts the following options:
13898 Set quality. This option defines the number of levels for averaging. It accepts
13899 an integer in the range 0-8. If set to @code{0}, the filter will have no
13900 effect. A value of @code{8} means the higher quality. For each increment of
13901 that value the speed drops by a factor of approximately 2. Default value is
13905 Force a constant quantization parameter. If not set, the filter will use the QP
13906 from the video stream (if available).
13909 @section vaguedenoiser
13911 Apply a wavelet based denoiser.
13913 It transforms each frame from the video input into the wavelet domain,
13914 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
13915 the obtained coefficients. It does an inverse wavelet transform after.
13916 Due to wavelet properties, it should give a nice smoothed result, and
13917 reduced noise, without blurring picture features.
13919 This filter accepts the following options:
13923 The filtering strength. The higher, the more filtered the video will be.
13924 Hard thresholding can use a higher threshold than soft thresholding
13925 before the video looks overfiltered.
13928 The filtering method the filter will use.
13930 It accepts the following values:
13933 All values under the threshold will be zeroed.
13936 All values under the threshold will be zeroed. All values above will be
13937 reduced by the threshold.
13940 Scales or nullifies coefficients - intermediary between (more) soft and
13941 (less) hard thresholding.
13945 Number of times, the wavelet will decompose the picture. Picture can't
13946 be decomposed beyond a particular point (typically, 8 for a 640x480
13947 frame - as 2^9 = 512 > 480)
13950 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
13953 A list of the planes to process. By default all planes are processed.
13956 @section vectorscope
13958 Display 2 color component values in the two dimensional graph (which is called
13961 This filter accepts the following options:
13965 Set vectorscope mode.
13967 It accepts the following values:
13970 Gray values are displayed on graph, higher brightness means more pixels have
13971 same component color value on location in graph. This is the default mode.
13974 Gray values are displayed on graph. Surrounding pixels values which are not
13975 present in video frame are drawn in gradient of 2 color components which are
13976 set by option @code{x} and @code{y}. The 3rd color component is static.
13979 Actual color components values present in video frame are displayed on graph.
13982 Similar as color2 but higher frequency of same values @code{x} and @code{y}
13983 on graph increases value of another color component, which is luminance by
13984 default values of @code{x} and @code{y}.
13987 Actual colors present in video frame are displayed on graph. If two different
13988 colors map to same position on graph then color with higher value of component
13989 not present in graph is picked.
13992 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
13993 component picked from radial gradient.
13997 Set which color component will be represented on X-axis. Default is @code{1}.
14000 Set which color component will be represented on Y-axis. Default is @code{2}.
14003 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
14004 of color component which represents frequency of (X, Y) location in graph.
14009 No envelope, this is default.
14012 Instant envelope, even darkest single pixel will be clearly highlighted.
14015 Hold maximum and minimum values presented in graph over time. This way you
14016 can still spot out of range values without constantly looking at vectorscope.
14019 Peak and instant envelope combined together.
14023 Set what kind of graticule to draw.
14031 Set graticule opacity.
14034 Set graticule flags.
14038 Draw graticule for white point.
14041 Draw graticule for black point.
14044 Draw color points short names.
14048 Set background opacity.
14050 @item lthreshold, l
14051 Set low threshold for color component not represented on X or Y axis.
14052 Values lower than this value will be ignored. Default is 0.
14053 Note this value is multiplied with actual max possible value one pixel component
14054 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14057 @item hthreshold, h
14058 Set high threshold for color component not represented on X or Y axis.
14059 Values higher than this value will be ignored. Default is 1.
14060 Note this value is multiplied with actual max possible value one pixel component
14061 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14062 is 0.9 * 255 = 230.
14064 @item colorspace, c
14065 Set what kind of colorspace to use when drawing graticule.
14074 @anchor{vidstabdetect}
14075 @section vidstabdetect
14077 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14078 @ref{vidstabtransform} for pass 2.
14080 This filter generates a file with relative translation and rotation
14081 transform information about subsequent frames, which is then used by
14082 the @ref{vidstabtransform} filter.
14084 To enable compilation of this filter you need to configure FFmpeg with
14085 @code{--enable-libvidstab}.
14087 This filter accepts the following options:
14091 Set the path to the file used to write the transforms information.
14092 Default value is @file{transforms.trf}.
14095 Set how shaky the video is and how quick the camera is. It accepts an
14096 integer in the range 1-10, a value of 1 means little shakiness, a
14097 value of 10 means strong shakiness. Default value is 5.
14100 Set the accuracy of the detection process. It must be a value in the
14101 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14102 accuracy. Default value is 15.
14105 Set stepsize of the search process. The region around minimum is
14106 scanned with 1 pixel resolution. Default value is 6.
14109 Set minimum contrast. Below this value a local measurement field is
14110 discarded. Must be a floating point value in the range 0-1. Default
14114 Set reference frame number for tripod mode.
14116 If enabled, the motion of the frames is compared to a reference frame
14117 in the filtered stream, identified by the specified number. The idea
14118 is to compensate all movements in a more-or-less static scene and keep
14119 the camera view absolutely still.
14121 If set to 0, it is disabled. The frames are counted starting from 1.
14124 Show fields and transforms in the resulting frames. It accepts an
14125 integer in the range 0-2. Default value is 0, which disables any
14129 @subsection Examples
14133 Use default values:
14139 Analyze strongly shaky movie and put the results in file
14140 @file{mytransforms.trf}:
14142 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14146 Visualize the result of internal transformations in the resulting
14149 vidstabdetect=show=1
14153 Analyze a video with medium shakiness using @command{ffmpeg}:
14155 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14159 @anchor{vidstabtransform}
14160 @section vidstabtransform
14162 Video stabilization/deshaking: pass 2 of 2,
14163 see @ref{vidstabdetect} for pass 1.
14165 Read a file with transform information for each frame and
14166 apply/compensate them. Together with the @ref{vidstabdetect}
14167 filter this can be used to deshake videos. See also
14168 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14169 the @ref{unsharp} filter, see below.
14171 To enable compilation of this filter you need to configure FFmpeg with
14172 @code{--enable-libvidstab}.
14174 @subsection Options
14178 Set path to the file used to read the transforms. Default value is
14179 @file{transforms.trf}.
14182 Set the number of frames (value*2 + 1) used for lowpass filtering the
14183 camera movements. Default value is 10.
14185 For example a number of 10 means that 21 frames are used (10 in the
14186 past and 10 in the future) to smoothen the motion in the video. A
14187 larger value leads to a smoother video, but limits the acceleration of
14188 the camera (pan/tilt movements). 0 is a special case where a static
14189 camera is simulated.
14192 Set the camera path optimization algorithm.
14194 Accepted values are:
14197 gaussian kernel low-pass filter on camera motion (default)
14199 averaging on transformations
14203 Set maximal number of pixels to translate frames. Default value is -1,
14207 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14208 value is -1, meaning no limit.
14211 Specify how to deal with borders that may be visible due to movement
14214 Available values are:
14217 keep image information from previous frame (default)
14219 fill the border black
14223 Invert transforms if set to 1. Default value is 0.
14226 Consider transforms as relative to previous frame if set to 1,
14227 absolute if set to 0. Default value is 0.
14230 Set percentage to zoom. A positive value will result in a zoom-in
14231 effect, a negative value in a zoom-out effect. Default value is 0 (no
14235 Set optimal zooming to avoid borders.
14237 Accepted values are:
14242 optimal static zoom value is determined (only very strong movements
14243 will lead to visible borders) (default)
14245 optimal adaptive zoom value is determined (no borders will be
14246 visible), see @option{zoomspeed}
14249 Note that the value given at zoom is added to the one calculated here.
14252 Set percent to zoom maximally each frame (enabled when
14253 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14257 Specify type of interpolation.
14259 Available values are:
14264 linear only horizontal
14266 linear in both directions (default)
14268 cubic in both directions (slow)
14272 Enable virtual tripod mode if set to 1, which is equivalent to
14273 @code{relative=0:smoothing=0}. Default value is 0.
14275 Use also @code{tripod} option of @ref{vidstabdetect}.
14278 Increase log verbosity if set to 1. Also the detected global motions
14279 are written to the temporary file @file{global_motions.trf}. Default
14283 @subsection Examples
14287 Use @command{ffmpeg} for a typical stabilization with default values:
14289 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14292 Note the use of the @ref{unsharp} filter which is always recommended.
14295 Zoom in a bit more and load transform data from a given file:
14297 vidstabtransform=zoom=5:input="mytransforms.trf"
14301 Smoothen the video even more:
14303 vidstabtransform=smoothing=30
14309 Flip the input video vertically.
14311 For example, to vertically flip a video with @command{ffmpeg}:
14313 ffmpeg -i in.avi -vf "vflip" out.avi
14319 Make or reverse a natural vignetting effect.
14321 The filter accepts the following options:
14325 Set lens angle expression as a number of radians.
14327 The value is clipped in the @code{[0,PI/2]} range.
14329 Default value: @code{"PI/5"}
14333 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14337 Set forward/backward mode.
14339 Available modes are:
14342 The larger the distance from the central point, the darker the image becomes.
14345 The larger the distance from the central point, the brighter the image becomes.
14346 This can be used to reverse a vignette effect, though there is no automatic
14347 detection to extract the lens @option{angle} and other settings (yet). It can
14348 also be used to create a burning effect.
14351 Default value is @samp{forward}.
14354 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14356 It accepts the following values:
14359 Evaluate expressions only once during the filter initialization.
14362 Evaluate expressions for each incoming frame. This is way slower than the
14363 @samp{init} mode since it requires all the scalers to be re-computed, but it
14364 allows advanced dynamic expressions.
14367 Default value is @samp{init}.
14370 Set dithering to reduce the circular banding effects. Default is @code{1}
14374 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14375 Setting this value to the SAR of the input will make a rectangular vignetting
14376 following the dimensions of the video.
14378 Default is @code{1/1}.
14381 @subsection Expressions
14383 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14384 following parameters.
14389 input width and height
14392 the number of input frame, starting from 0
14395 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14396 @var{TB} units, NAN if undefined
14399 frame rate of the input video, NAN if the input frame rate is unknown
14402 the PTS (Presentation TimeStamp) of the filtered video frame,
14403 expressed in seconds, NAN if undefined
14406 time base of the input video
14410 @subsection Examples
14414 Apply simple strong vignetting effect:
14420 Make a flickering vignetting:
14422 vignette='PI/4+random(1)*PI/50':eval=frame
14428 Stack input videos vertically.
14430 All streams must be of same pixel format and of same width.
14432 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14433 to create same output.
14435 The filter accept the following option:
14439 Set number of input streams. Default is 2.
14442 If set to 1, force the output to terminate when the shortest input
14443 terminates. Default value is 0.
14448 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14449 Deinterlacing Filter").
14451 Based on the process described by Martin Weston for BBC R&D, and
14452 implemented based on the de-interlace algorithm written by Jim
14453 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14454 uses filter coefficients calculated by BBC R&D.
14456 There are two sets of filter coefficients, so called "simple":
14457 and "complex". Which set of filter coefficients is used can
14458 be set by passing an optional parameter:
14462 Set the interlacing filter coefficients. Accepts one of the following values:
14466 Simple filter coefficient set.
14468 More-complex filter coefficient set.
14470 Default value is @samp{complex}.
14473 Specify which frames to deinterlace. Accept one of the following values:
14477 Deinterlace all frames,
14479 Only deinterlace frames marked as interlaced.
14482 Default value is @samp{all}.
14486 Video waveform monitor.
14488 The waveform monitor plots color component intensity. By default luminance
14489 only. Each column of the waveform corresponds to a column of pixels in the
14492 It accepts the following options:
14496 Can be either @code{row}, or @code{column}. Default is @code{column}.
14497 In row mode, the graph on the left side represents color component value 0 and
14498 the right side represents value = 255. In column mode, the top side represents
14499 color component value = 0 and bottom side represents value = 255.
14502 Set intensity. Smaller values are useful to find out how many values of the same
14503 luminance are distributed across input rows/columns.
14504 Default value is @code{0.04}. Allowed range is [0, 1].
14507 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14508 In mirrored mode, higher values will be represented on the left
14509 side for @code{row} mode and at the top for @code{column} mode. Default is
14510 @code{1} (mirrored).
14514 It accepts the following values:
14517 Presents information identical to that in the @code{parade}, except
14518 that the graphs representing color components are superimposed directly
14521 This display mode makes it easier to spot relative differences or similarities
14522 in overlapping areas of the color components that are supposed to be identical,
14523 such as neutral whites, grays, or blacks.
14526 Display separate graph for the color components side by side in
14527 @code{row} mode or one below the other in @code{column} mode.
14530 Display separate graph for the color components side by side in
14531 @code{column} mode or one below the other in @code{row} mode.
14533 Using this display mode makes it easy to spot color casts in the highlights
14534 and shadows of an image, by comparing the contours of the top and the bottom
14535 graphs of each waveform. Since whites, grays, and blacks are characterized
14536 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14537 should display three waveforms of roughly equal width/height. If not, the
14538 correction is easy to perform by making level adjustments the three waveforms.
14540 Default is @code{stack}.
14542 @item components, c
14543 Set which color components to display. Default is 1, which means only luminance
14544 or red color component if input is in RGB colorspace. If is set for example to
14545 7 it will display all 3 (if) available color components.
14550 No envelope, this is default.
14553 Instant envelope, minimum and maximum values presented in graph will be easily
14554 visible even with small @code{step} value.
14557 Hold minimum and maximum values presented in graph across time. This way you
14558 can still spot out of range values without constantly looking at waveforms.
14561 Peak and instant envelope combined together.
14567 No filtering, this is default.
14570 Luma and chroma combined together.
14573 Similar as above, but shows difference between blue and red chroma.
14576 Displays only chroma.
14579 Displays actual color value on waveform.
14582 Similar as above, but with luma showing frequency of chroma values.
14586 Set which graticule to display.
14590 Do not display graticule.
14593 Display green graticule showing legal broadcast ranges.
14597 Set graticule opacity.
14600 Set graticule flags.
14604 Draw numbers above lines. By default enabled.
14607 Draw dots instead of lines.
14611 Set scale used for displaying graticule.
14618 Default is digital.
14621 Set background opacity.
14626 The @code{weave} takes a field-based video input and join
14627 each two sequential fields into single frame, producing a new double
14628 height clip with half the frame rate and half the frame count.
14630 It accepts the following option:
14634 Set first field. Available values are:
14638 Set the frame as top-field-first.
14641 Set the frame as bottom-field-first.
14645 @subsection Examples
14649 Interlace video using @ref{select} and @ref{separatefields} filter:
14651 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14656 Apply the xBR high-quality magnification filter which is designed for pixel
14657 art. It follows a set of edge-detection rules, see
14658 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14660 It accepts the following option:
14664 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14665 @code{3xBR} and @code{4} for @code{4xBR}.
14666 Default is @code{3}.
14672 Deinterlace the input video ("yadif" means "yet another deinterlacing
14675 It accepts the following parameters:
14681 The interlacing mode to adopt. It accepts one of the following values:
14684 @item 0, send_frame
14685 Output one frame for each frame.
14686 @item 1, send_field
14687 Output one frame for each field.
14688 @item 2, send_frame_nospatial
14689 Like @code{send_frame}, but it skips the spatial interlacing check.
14690 @item 3, send_field_nospatial
14691 Like @code{send_field}, but it skips the spatial interlacing check.
14694 The default value is @code{send_frame}.
14697 The picture field parity assumed for the input interlaced video. It accepts one
14698 of the following values:
14702 Assume the top field is first.
14704 Assume the bottom field is first.
14706 Enable automatic detection of field parity.
14709 The default value is @code{auto}.
14710 If the interlacing is unknown or the decoder does not export this information,
14711 top field first will be assumed.
14714 Specify which frames to deinterlace. Accept one of the following
14719 Deinterlace all frames.
14720 @item 1, interlaced
14721 Only deinterlace frames marked as interlaced.
14724 The default value is @code{all}.
14729 Apply Zoom & Pan effect.
14731 This filter accepts the following options:
14735 Set the zoom expression. Default is 1.
14739 Set the x and y expression. Default is 0.
14742 Set the duration expression in number of frames.
14743 This sets for how many number of frames effect will last for
14744 single input image.
14747 Set the output image size, default is 'hd720'.
14750 Set the output frame rate, default is '25'.
14753 Each expression can contain the following constants:
14772 Output frame count.
14776 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14777 for current input frame.
14781 'x' and 'y' of last output frame of previous input frame or 0 when there was
14782 not yet such frame (first input frame).
14785 Last calculated zoom from 'z' expression for current input frame.
14788 Last calculated zoom of last output frame of previous input frame.
14791 Number of output frames for current input frame. Calculated from 'd' expression
14792 for each input frame.
14795 number of output frames created for previous input frame
14798 Rational number: input width / input height
14801 sample aspect ratio
14804 display aspect ratio
14808 @subsection Examples
14812 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14814 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
14818 Zoom-in up to 1.5 and pan always at center of picture:
14820 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14824 Same as above but without pausing:
14826 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14831 Scale (resize) the input video, using the z.lib library:
14832 https://github.com/sekrit-twc/zimg.
14834 The zscale filter forces the output display aspect ratio to be the same
14835 as the input, by changing the output sample aspect ratio.
14837 If the input image format is different from the format requested by
14838 the next filter, the zscale filter will convert the input to the
14841 @subsection Options
14842 The filter accepts the following options.
14847 Set the output video dimension expression. Default value is the input
14850 If the @var{width} or @var{w} is 0, the input width is used for the output.
14851 If the @var{height} or @var{h} is 0, the input height is used for the output.
14853 If one of the values is -1, the zscale filter will use a value that
14854 maintains the aspect ratio of the input image, calculated from the
14855 other specified dimension. If both of them are -1, the input size is
14858 If one of the values is -n with n > 1, the zscale filter will also use a value
14859 that maintains the aspect ratio of the input image, calculated from the other
14860 specified dimension. After that it will, however, make sure that the calculated
14861 dimension is divisible by n and adjust the value if necessary.
14863 See below for the list of accepted constants for use in the dimension
14867 Set the video size. For the syntax of this option, check the
14868 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14871 Set the dither type.
14873 Possible values are:
14878 @item error_diffusion
14884 Set the resize filter type.
14886 Possible values are:
14896 Default is bilinear.
14899 Set the color range.
14901 Possible values are:
14908 Default is same as input.
14911 Set the color primaries.
14913 Possible values are:
14923 Default is same as input.
14926 Set the transfer characteristics.
14928 Possible values are:
14942 Default is same as input.
14945 Set the colorspace matrix.
14947 Possible value are:
14958 Default is same as input.
14961 Set the input color range.
14963 Possible values are:
14970 Default is same as input.
14972 @item primariesin, pin
14973 Set the input color primaries.
14975 Possible values are:
14985 Default is same as input.
14987 @item transferin, tin
14988 Set the input transfer characteristics.
14990 Possible values are:
15001 Default is same as input.
15003 @item matrixin, min
15004 Set the input colorspace matrix.
15006 Possible value are:
15018 Set the output chroma location.
15020 Possible values are:
15031 @item chromalin, cin
15032 Set the input chroma location.
15034 Possible values are:
15046 Set the nominal peak luminance.
15049 The values of the @option{w} and @option{h} options are expressions
15050 containing the following constants:
15055 The input width and height
15059 These are the same as @var{in_w} and @var{in_h}.
15063 The output (scaled) width and height
15067 These are the same as @var{out_w} and @var{out_h}
15070 The same as @var{iw} / @var{ih}
15073 input sample aspect ratio
15076 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15080 horizontal and vertical input chroma subsample values. For example for the
15081 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15085 horizontal and vertical output chroma subsample values. For example for the
15086 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15092 @c man end VIDEO FILTERS
15094 @chapter Video Sources
15095 @c man begin VIDEO SOURCES
15097 Below is a description of the currently available video sources.
15101 Buffer video frames, and make them available to the filter chain.
15103 This source is mainly intended for a programmatic use, in particular
15104 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15106 It accepts the following parameters:
15111 Specify the size (width and height) of the buffered video frames. For the
15112 syntax of this option, check the
15113 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15116 The input video width.
15119 The input video height.
15122 A string representing the pixel format of the buffered video frames.
15123 It may be a number corresponding to a pixel format, or a pixel format
15127 Specify the timebase assumed by the timestamps of the buffered frames.
15130 Specify the frame rate expected for the video stream.
15132 @item pixel_aspect, sar
15133 The sample (pixel) aspect ratio of the input video.
15136 Specify the optional parameters to be used for the scale filter which
15137 is automatically inserted when an input change is detected in the
15138 input size or format.
15140 @item hw_frames_ctx
15141 When using a hardware pixel format, this should be a reference to an
15142 AVHWFramesContext describing input frames.
15147 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15150 will instruct the source to accept video frames with size 320x240 and
15151 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15152 square pixels (1:1 sample aspect ratio).
15153 Since the pixel format with name "yuv410p" corresponds to the number 6
15154 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15155 this example corresponds to:
15157 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15160 Alternatively, the options can be specified as a flat string, but this
15161 syntax is deprecated:
15163 @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}]
15167 Create a pattern generated by an elementary cellular automaton.
15169 The initial state of the cellular automaton can be defined through the
15170 @option{filename} and @option{pattern} options. If such options are
15171 not specified an initial state is created randomly.
15173 At each new frame a new row in the video is filled with the result of
15174 the cellular automaton next generation. The behavior when the whole
15175 frame is filled is defined by the @option{scroll} option.
15177 This source accepts the following options:
15181 Read the initial cellular automaton state, i.e. the starting row, from
15182 the specified file.
15183 In the file, each non-whitespace character is considered an alive
15184 cell, a newline will terminate the row, and further characters in the
15185 file will be ignored.
15188 Read the initial cellular automaton state, i.e. the starting row, from
15189 the specified string.
15191 Each non-whitespace character in the string is considered an alive
15192 cell, a newline will terminate the row, and further characters in the
15193 string will be ignored.
15196 Set the video rate, that is the number of frames generated per second.
15199 @item random_fill_ratio, ratio
15200 Set the random fill ratio for the initial cellular automaton row. It
15201 is a floating point number value ranging from 0 to 1, defaults to
15204 This option is ignored when a file or a pattern is specified.
15206 @item random_seed, seed
15207 Set the seed for filling randomly the initial row, must be an integer
15208 included between 0 and UINT32_MAX. If not specified, or if explicitly
15209 set to -1, the filter will try to use a good random seed on a best
15213 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15214 Default value is 110.
15217 Set the size of the output video. For the syntax of this option, check the
15218 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15220 If @option{filename} or @option{pattern} is specified, the size is set
15221 by default to the width of the specified initial state row, and the
15222 height is set to @var{width} * PHI.
15224 If @option{size} is set, it must contain the width of the specified
15225 pattern string, and the specified pattern will be centered in the
15228 If a filename or a pattern string is not specified, the size value
15229 defaults to "320x518" (used for a randomly generated initial state).
15232 If set to 1, scroll the output upward when all the rows in the output
15233 have been already filled. If set to 0, the new generated row will be
15234 written over the top row just after the bottom row is filled.
15237 @item start_full, full
15238 If set to 1, completely fill the output with generated rows before
15239 outputting the first frame.
15240 This is the default behavior, for disabling set the value to 0.
15243 If set to 1, stitch the left and right row edges together.
15244 This is the default behavior, for disabling set the value to 0.
15247 @subsection Examples
15251 Read the initial state from @file{pattern}, and specify an output of
15254 cellauto=f=pattern:s=200x400
15258 Generate a random initial row with a width of 200 cells, with a fill
15261 cellauto=ratio=2/3:s=200x200
15265 Create a pattern generated by rule 18 starting by a single alive cell
15266 centered on an initial row with width 100:
15268 cellauto=p=@@:s=100x400:full=0:rule=18
15272 Specify a more elaborated initial pattern:
15274 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15279 @anchor{coreimagesrc}
15280 @section coreimagesrc
15281 Video source generated on GPU using Apple's CoreImage API on OSX.
15283 This video source is a specialized version of the @ref{coreimage} video filter.
15284 Use a core image generator at the beginning of the applied filterchain to
15285 generate the content.
15287 The coreimagesrc video source accepts the following options:
15289 @item list_generators
15290 List all available generators along with all their respective options as well as
15291 possible minimum and maximum values along with the default values.
15293 list_generators=true
15297 Specify the size of the sourced video. For the syntax of this option, check the
15298 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15299 The default value is @code{320x240}.
15302 Specify the frame rate of the sourced video, as the number of frames
15303 generated per second. It has to be a string in the format
15304 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15305 number or a valid video frame rate abbreviation. The default value is
15309 Set the sample aspect ratio of the sourced video.
15312 Set the duration of the sourced video. See
15313 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15314 for the accepted syntax.
15316 If not specified, or the expressed duration is negative, the video is
15317 supposed to be generated forever.
15320 Additionally, all options of the @ref{coreimage} video filter are accepted.
15321 A complete filterchain can be used for further processing of the
15322 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15323 and examples for details.
15325 @subsection Examples
15330 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15331 given as complete and escaped command-line for Apple's standard bash shell:
15333 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15335 This example is equivalent to the QRCode example of @ref{coreimage} without the
15336 need for a nullsrc video source.
15340 @section mandelbrot
15342 Generate a Mandelbrot set fractal, and progressively zoom towards the
15343 point specified with @var{start_x} and @var{start_y}.
15345 This source accepts the following options:
15350 Set the terminal pts value. Default value is 400.
15353 Set the terminal scale value.
15354 Must be a floating point value. Default value is 0.3.
15357 Set the inner coloring mode, that is the algorithm used to draw the
15358 Mandelbrot fractal internal region.
15360 It shall assume one of the following values:
15365 Show time until convergence.
15367 Set color based on point closest to the origin of the iterations.
15372 Default value is @var{mincol}.
15375 Set the bailout value. Default value is 10.0.
15378 Set the maximum of iterations performed by the rendering
15379 algorithm. Default value is 7189.
15382 Set outer coloring mode.
15383 It shall assume one of following values:
15385 @item iteration_count
15386 Set iteration cound mode.
15387 @item normalized_iteration_count
15388 set normalized iteration count mode.
15390 Default value is @var{normalized_iteration_count}.
15393 Set frame rate, expressed as number of frames per second. Default
15397 Set frame size. For the syntax of this option, check the "Video
15398 size" section in the ffmpeg-utils manual. Default value is "640x480".
15401 Set the initial scale value. Default value is 3.0.
15404 Set the initial x position. Must be a floating point value between
15405 -100 and 100. Default value is -0.743643887037158704752191506114774.
15408 Set the initial y position. Must be a floating point value between
15409 -100 and 100. Default value is -0.131825904205311970493132056385139.
15414 Generate various test patterns, as generated by the MPlayer test filter.
15416 The size of the generated video is fixed, and is 256x256.
15417 This source is useful in particular for testing encoding features.
15419 This source accepts the following options:
15424 Specify the frame rate of the sourced video, as the number of frames
15425 generated per second. It has to be a string in the format
15426 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15427 number or a valid video frame rate abbreviation. The default value is
15431 Set the duration of the sourced video. See
15432 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15433 for the accepted syntax.
15435 If not specified, or the expressed duration is negative, the video is
15436 supposed to be generated forever.
15440 Set the number or the name of the test to perform. Supported tests are:
15456 Default value is "all", which will cycle through the list of all tests.
15461 mptestsrc=t=dc_luma
15464 will generate a "dc_luma" test pattern.
15466 @section frei0r_src
15468 Provide a frei0r source.
15470 To enable compilation of this filter you need to install the frei0r
15471 header and configure FFmpeg with @code{--enable-frei0r}.
15473 This source accepts the following parameters:
15478 The size of the video to generate. For the syntax of this option, check the
15479 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15482 The framerate of the generated video. It may be a string of the form
15483 @var{num}/@var{den} or a frame rate abbreviation.
15486 The name to the frei0r source to load. For more information regarding frei0r and
15487 how to set the parameters, read the @ref{frei0r} section in the video filters
15490 @item filter_params
15491 A '|'-separated list of parameters to pass to the frei0r source.
15495 For example, to generate a frei0r partik0l source with size 200x200
15496 and frame rate 10 which is overlaid on the overlay filter main input:
15498 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15503 Generate a life pattern.
15505 This source is based on a generalization of John Conway's life game.
15507 The sourced input represents a life grid, each pixel represents a cell
15508 which can be in one of two possible states, alive or dead. Every cell
15509 interacts with its eight neighbours, which are the cells that are
15510 horizontally, vertically, or diagonally adjacent.
15512 At each interaction the grid evolves according to the adopted rule,
15513 which specifies the number of neighbor alive cells which will make a
15514 cell stay alive or born. The @option{rule} option allows one to specify
15517 This source accepts the following options:
15521 Set the file from which to read the initial grid state. In the file,
15522 each non-whitespace character is considered an alive cell, and newline
15523 is used to delimit the end of each row.
15525 If this option is not specified, the initial grid is generated
15529 Set the video rate, that is the number of frames generated per second.
15532 @item random_fill_ratio, ratio
15533 Set the random fill ratio for the initial random grid. It is a
15534 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15535 It is ignored when a file is specified.
15537 @item random_seed, seed
15538 Set the seed for filling the initial random grid, must be an integer
15539 included between 0 and UINT32_MAX. If not specified, or if explicitly
15540 set to -1, the filter will try to use a good random seed on a best
15546 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15547 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15548 @var{NS} specifies the number of alive neighbor cells which make a
15549 live cell stay alive, and @var{NB} the number of alive neighbor cells
15550 which make a dead cell to become alive (i.e. to "born").
15551 "s" and "b" can be used in place of "S" and "B", respectively.
15553 Alternatively a rule can be specified by an 18-bits integer. The 9
15554 high order bits are used to encode the next cell state if it is alive
15555 for each number of neighbor alive cells, the low order bits specify
15556 the rule for "borning" new cells. Higher order bits encode for an
15557 higher number of neighbor cells.
15558 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15559 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15561 Default value is "S23/B3", which is the original Conway's game of life
15562 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15563 cells, and will born a new cell if there are three alive cells around
15567 Set the size of the output video. For the syntax of this option, check the
15568 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15570 If @option{filename} is specified, the size is set by default to the
15571 same size of the input file. If @option{size} is set, it must contain
15572 the size specified in the input file, and the initial grid defined in
15573 that file is centered in the larger resulting area.
15575 If a filename is not specified, the size value defaults to "320x240"
15576 (used for a randomly generated initial grid).
15579 If set to 1, stitch the left and right grid edges together, and the
15580 top and bottom edges also. Defaults to 1.
15583 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15584 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15585 value from 0 to 255.
15588 Set the color of living (or new born) cells.
15591 Set the color of dead cells. If @option{mold} is set, this is the first color
15592 used to represent a dead cell.
15595 Set mold color, for definitely dead and moldy cells.
15597 For the syntax of these 3 color options, check the "Color" section in the
15598 ffmpeg-utils manual.
15601 @subsection Examples
15605 Read a grid from @file{pattern}, and center it on a grid of size
15608 life=f=pattern:s=300x300
15612 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15614 life=ratio=2/3:s=200x200
15618 Specify a custom rule for evolving a randomly generated grid:
15624 Full example with slow death effect (mold) using @command{ffplay}:
15626 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15633 @anchor{haldclutsrc}
15635 @anchor{rgbtestsrc}
15637 @anchor{smptehdbars}
15640 @anchor{yuvtestsrc}
15641 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15643 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15645 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15647 The @code{color} source provides an uniformly colored input.
15649 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15650 @ref{haldclut} filter.
15652 The @code{nullsrc} source returns unprocessed video frames. It is
15653 mainly useful to be employed in analysis / debugging tools, or as the
15654 source for filters which ignore the input data.
15656 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15657 detecting RGB vs BGR issues. You should see a red, green and blue
15658 stripe from top to bottom.
15660 The @code{smptebars} source generates a color bars pattern, based on
15661 the SMPTE Engineering Guideline EG 1-1990.
15663 The @code{smptehdbars} source generates a color bars pattern, based on
15664 the SMPTE RP 219-2002.
15666 The @code{testsrc} source generates a test video pattern, showing a
15667 color pattern, a scrolling gradient and a timestamp. This is mainly
15668 intended for testing purposes.
15670 The @code{testsrc2} source is similar to testsrc, but supports more
15671 pixel formats instead of just @code{rgb24}. This allows using it as an
15672 input for other tests without requiring a format conversion.
15674 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15675 see a y, cb and cr stripe from top to bottom.
15677 The sources accept the following parameters:
15682 Specify the color of the source, only available in the @code{color}
15683 source. For the syntax of this option, check the "Color" section in the
15684 ffmpeg-utils manual.
15687 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15688 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15689 pixels to be used as identity matrix for 3D lookup tables. Each component is
15690 coded on a @code{1/(N*N)} scale.
15693 Specify the size of the sourced video. For the syntax of this option, check the
15694 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15695 The default value is @code{320x240}.
15697 This option is not available with the @code{haldclutsrc} filter.
15700 Specify the frame rate of the sourced video, as the number of frames
15701 generated per second. It has to be a string in the format
15702 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15703 number or a valid video frame rate abbreviation. The default value is
15707 Set the sample aspect ratio of the sourced video.
15710 Set the duration of the sourced video. See
15711 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15712 for the accepted syntax.
15714 If not specified, or the expressed duration is negative, the video is
15715 supposed to be generated forever.
15718 Set the number of decimals to show in the timestamp, only available in the
15719 @code{testsrc} source.
15721 The displayed timestamp value will correspond to the original
15722 timestamp value multiplied by the power of 10 of the specified
15723 value. Default value is 0.
15726 For example the following:
15728 testsrc=duration=5.3:size=qcif:rate=10
15731 will generate a video with a duration of 5.3 seconds, with size
15732 176x144 and a frame rate of 10 frames per second.
15734 The following graph description will generate a red source
15735 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15738 color=c=red@@0.2:s=qcif:r=10
15741 If the input content is to be ignored, @code{nullsrc} can be used. The
15742 following command generates noise in the luminance plane by employing
15743 the @code{geq} filter:
15745 nullsrc=s=256x256, geq=random(1)*255:128:128
15748 @subsection Commands
15750 The @code{color} source supports the following commands:
15754 Set the color of the created image. Accepts the same syntax of the
15755 corresponding @option{color} option.
15758 @c man end VIDEO SOURCES
15760 @chapter Video Sinks
15761 @c man begin VIDEO SINKS
15763 Below is a description of the currently available video sinks.
15765 @section buffersink
15767 Buffer video frames, and make them available to the end of the filter
15770 This sink is mainly intended for programmatic use, in particular
15771 through the interface defined in @file{libavfilter/buffersink.h}
15772 or the options system.
15774 It accepts a pointer to an AVBufferSinkContext structure, which
15775 defines the incoming buffers' formats, to be passed as the opaque
15776 parameter to @code{avfilter_init_filter} for initialization.
15780 Null video sink: do absolutely nothing with the input video. It is
15781 mainly useful as a template and for use in analysis / debugging
15784 @c man end VIDEO SINKS
15786 @chapter Multimedia Filters
15787 @c man begin MULTIMEDIA FILTERS
15789 Below is a description of the currently available multimedia filters.
15793 Convert input audio to a video output, displaying the audio bit scope.
15795 The filter accepts the following options:
15799 Set frame rate, expressed as number of frames per second. Default
15803 Specify the video size for the output. For the syntax of this option, check the
15804 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15805 Default value is @code{1024x256}.
15808 Specify list of colors separated by space or by '|' which will be used to
15809 draw channels. Unrecognized or missing colors will be replaced
15813 @section ahistogram
15815 Convert input audio to a video output, displaying the volume histogram.
15817 The filter accepts the following options:
15821 Specify how histogram is calculated.
15823 It accepts the following values:
15826 Use single histogram for all channels.
15828 Use separate histogram for each channel.
15830 Default is @code{single}.
15833 Set frame rate, expressed as number of frames per second. Default
15837 Specify the video size for the output. For the syntax of this option, check the
15838 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15839 Default value is @code{hd720}.
15844 It accepts the following values:
15855 reverse logarithmic
15857 Default is @code{log}.
15860 Set amplitude scale.
15862 It accepts the following values:
15869 Default is @code{log}.
15872 Set how much frames to accumulate in histogram.
15873 Defauls is 1. Setting this to -1 accumulates all frames.
15876 Set histogram ratio of window height.
15879 Set sonogram sliding.
15881 It accepts the following values:
15884 replace old rows with new ones.
15886 scroll from top to bottom.
15888 Default is @code{replace}.
15891 @section aphasemeter
15893 Convert input audio to a video output, displaying the audio phase.
15895 The filter accepts the following options:
15899 Set the output frame rate. Default value is @code{25}.
15902 Set the video size for the output. For the syntax of this option, check the
15903 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15904 Default value is @code{800x400}.
15909 Specify the red, green, blue contrast. Default values are @code{2},
15910 @code{7} and @code{1}.
15911 Allowed range is @code{[0, 255]}.
15914 Set color which will be used for drawing median phase. If color is
15915 @code{none} which is default, no median phase value will be drawn.
15918 Enable video output. Default is enabled.
15921 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
15922 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
15923 The @code{-1} means left and right channels are completely out of phase and
15924 @code{1} means channels are in phase.
15926 @section avectorscope
15928 Convert input audio to a video output, representing the audio vector
15931 The filter is used to measure the difference between channels of stereo
15932 audio stream. A monoaural signal, consisting of identical left and right
15933 signal, results in straight vertical line. Any stereo separation is visible
15934 as a deviation from this line, creating a Lissajous figure.
15935 If the straight (or deviation from it) but horizontal line appears this
15936 indicates that the left and right channels are out of phase.
15938 The filter accepts the following options:
15942 Set the vectorscope mode.
15944 Available values are:
15947 Lissajous rotated by 45 degrees.
15950 Same as above but not rotated.
15953 Shape resembling half of circle.
15956 Default value is @samp{lissajous}.
15959 Set the video size for the output. For the syntax of this option, check the
15960 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15961 Default value is @code{400x400}.
15964 Set the output frame rate. Default value is @code{25}.
15970 Specify the red, green, blue and alpha contrast. Default values are @code{40},
15971 @code{160}, @code{80} and @code{255}.
15972 Allowed range is @code{[0, 255]}.
15978 Specify the red, green, blue and alpha fade. Default values are @code{15},
15979 @code{10}, @code{5} and @code{5}.
15980 Allowed range is @code{[0, 255]}.
15983 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
15986 Set the vectorscope drawing mode.
15988 Available values are:
15991 Draw dot for each sample.
15994 Draw line between previous and current sample.
15997 Default value is @samp{dot}.
16000 Specify amplitude scale of audio samples.
16002 Available values are:
16019 @subsection Examples
16023 Complete example using @command{ffplay}:
16025 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16026 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16030 @section bench, abench
16032 Benchmark part of a filtergraph.
16034 The filter accepts the following options:
16038 Start or stop a timer.
16040 Available values are:
16043 Get the current time, set it as frame metadata (using the key
16044 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
16047 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16048 the input frame metadata to get the time difference. Time difference, average,
16049 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16050 @code{min}) are then printed. The timestamps are expressed in seconds.
16054 @subsection Examples
16058 Benchmark @ref{selectivecolor} filter:
16060 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16066 Concatenate audio and video streams, joining them together one after the
16069 The filter works on segments of synchronized video and audio streams. All
16070 segments must have the same number of streams of each type, and that will
16071 also be the number of streams at output.
16073 The filter accepts the following options:
16078 Set the number of segments. Default is 2.
16081 Set the number of output video streams, that is also the number of video
16082 streams in each segment. Default is 1.
16085 Set the number of output audio streams, that is also the number of audio
16086 streams in each segment. Default is 0.
16089 Activate unsafe mode: do not fail if segments have a different format.
16093 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16094 @var{a} audio outputs.
16096 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16097 segment, in the same order as the outputs, then the inputs for the second
16100 Related streams do not always have exactly the same duration, for various
16101 reasons including codec frame size or sloppy authoring. For that reason,
16102 related synchronized streams (e.g. a video and its audio track) should be
16103 concatenated at once. The concat filter will use the duration of the longest
16104 stream in each segment (except the last one), and if necessary pad shorter
16105 audio streams with silence.
16107 For this filter to work correctly, all segments must start at timestamp 0.
16109 All corresponding streams must have the same parameters in all segments; the
16110 filtering system will automatically select a common pixel format for video
16111 streams, and a common sample format, sample rate and channel layout for
16112 audio streams, but other settings, such as resolution, must be converted
16113 explicitly by the user.
16115 Different frame rates are acceptable but will result in variable frame rate
16116 at output; be sure to configure the output file to handle it.
16118 @subsection Examples
16122 Concatenate an opening, an episode and an ending, all in bilingual version
16123 (video in stream 0, audio in streams 1 and 2):
16125 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16126 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16127 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16128 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16132 Concatenate two parts, handling audio and video separately, using the
16133 (a)movie sources, and adjusting the resolution:
16135 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16136 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16137 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16139 Note that a desync will happen at the stitch if the audio and video streams
16140 do not have exactly the same duration in the first file.
16144 @section drawgraph, adrawgraph
16146 Draw a graph using input video or audio metadata.
16148 It accepts the following parameters:
16152 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16155 Set 1st foreground color expression.
16158 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16161 Set 2nd foreground color expression.
16164 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16167 Set 3rd foreground color expression.
16170 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16173 Set 4th foreground color expression.
16176 Set minimal value of metadata value.
16179 Set maximal value of metadata value.
16182 Set graph background color. Default is white.
16187 Available values for mode is:
16194 Default is @code{line}.
16199 Available values for slide is:
16202 Draw new frame when right border is reached.
16205 Replace old columns with new ones.
16208 Scroll from right to left.
16211 Scroll from left to right.
16214 Draw single picture.
16217 Default is @code{frame}.
16220 Set size of graph video. For the syntax of this option, check the
16221 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16222 The default value is @code{900x256}.
16224 The foreground color expressions can use the following variables:
16227 Minimal value of metadata value.
16230 Maximal value of metadata value.
16233 Current metadata key value.
16236 The color is defined as 0xAABBGGRR.
16239 Example using metadata from @ref{signalstats} filter:
16241 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16244 Example using metadata from @ref{ebur128} filter:
16246 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16252 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16253 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16254 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16255 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16257 The filter also has a video output (see the @var{video} option) with a real
16258 time graph to observe the loudness evolution. The graphic contains the logged
16259 message mentioned above, so it is not printed anymore when this option is set,
16260 unless the verbose logging is set. The main graphing area contains the
16261 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16262 the momentary loudness (400 milliseconds).
16264 More information about the Loudness Recommendation EBU R128 on
16265 @url{http://tech.ebu.ch/loudness}.
16267 The filter accepts the following options:
16272 Activate the video output. The audio stream is passed unchanged whether this
16273 option is set or no. The video stream will be the first output stream if
16274 activated. Default is @code{0}.
16277 Set the video size. This option is for video only. For the syntax of this
16279 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16280 Default and minimum resolution is @code{640x480}.
16283 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16284 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16285 other integer value between this range is allowed.
16288 Set metadata injection. If set to @code{1}, the audio input will be segmented
16289 into 100ms output frames, each of them containing various loudness information
16290 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16292 Default is @code{0}.
16295 Force the frame logging level.
16297 Available values are:
16300 information logging level
16302 verbose logging level
16305 By default, the logging level is set to @var{info}. If the @option{video} or
16306 the @option{metadata} options are set, it switches to @var{verbose}.
16311 Available modes can be cumulated (the option is a @code{flag} type). Possible
16315 Disable any peak mode (default).
16317 Enable sample-peak mode.
16319 Simple peak mode looking for the higher sample value. It logs a message
16320 for sample-peak (identified by @code{SPK}).
16322 Enable true-peak mode.
16324 If enabled, the peak lookup is done on an over-sampled version of the input
16325 stream for better peak accuracy. It logs a message for true-peak.
16326 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16327 This mode requires a build with @code{libswresample}.
16331 Treat mono input files as "dual mono". If a mono file is intended for playback
16332 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16333 If set to @code{true}, this option will compensate for this effect.
16334 Multi-channel input files are not affected by this option.
16337 Set a specific pan law to be used for the measurement of dual mono files.
16338 This parameter is optional, and has a default value of -3.01dB.
16341 @subsection Examples
16345 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16347 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16351 Run an analysis with @command{ffmpeg}:
16353 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16357 @section interleave, ainterleave
16359 Temporally interleave frames from several inputs.
16361 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16363 These filters read frames from several inputs and send the oldest
16364 queued frame to the output.
16366 Input streams must have well defined, monotonically increasing frame
16369 In order to submit one frame to output, these filters need to enqueue
16370 at least one frame for each input, so they cannot work in case one
16371 input is not yet terminated and will not receive incoming frames.
16373 For example consider the case when one input is a @code{select} filter
16374 which always drops input frames. The @code{interleave} filter will keep
16375 reading from that input, but it will never be able to send new frames
16376 to output until the input sends an end-of-stream signal.
16378 Also, depending on inputs synchronization, the filters will drop
16379 frames in case one input receives more frames than the other ones, and
16380 the queue is already filled.
16382 These filters accept the following options:
16386 Set the number of different inputs, it is 2 by default.
16389 @subsection Examples
16393 Interleave frames belonging to different streams using @command{ffmpeg}:
16395 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16399 Add flickering blur effect:
16401 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16405 @section metadata, ametadata
16407 Manipulate frame metadata.
16409 This filter accepts the following options:
16413 Set mode of operation of the filter.
16415 Can be one of the following:
16419 If both @code{value} and @code{key} is set, select frames
16420 which have such metadata. If only @code{key} is set, select
16421 every frame that has such key in metadata.
16424 Add new metadata @code{key} and @code{value}. If key is already available
16428 Modify value of already present key.
16431 If @code{value} is set, delete only keys that have such value.
16432 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16436 Print key and its value if metadata was found. If @code{key} is not set print all
16437 metadata values available in frame.
16441 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16444 Set metadata value which will be used. This option is mandatory for
16445 @code{modify} and @code{add} mode.
16448 Which function to use when comparing metadata value and @code{value}.
16450 Can be one of following:
16454 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16457 Values are interpreted as strings, returns true if metadata value starts with
16458 the @code{value} option string.
16461 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16464 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16467 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16470 Values are interpreted as floats, returns true if expression from option @code{expr}
16475 Set expression which is used when @code{function} is set to @code{expr}.
16476 The expression is evaluated through the eval API and can contain the following
16481 Float representation of @code{value} from metadata key.
16484 Float representation of @code{value} as supplied by user in @code{value} option.
16488 If specified in @code{print} mode, output is written to the named file. Instead of
16489 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16490 for standard output. If @code{file} option is not set, output is written to the log
16491 with AV_LOG_INFO loglevel.
16495 @subsection Examples
16499 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16502 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16505 Print silencedetect output to file @file{metadata.txt}.
16507 silencedetect,ametadata=mode=print:file=metadata.txt
16510 Direct all metadata to a pipe with file descriptor 4.
16512 metadata=mode=print:file='pipe\:4'
16516 @section perms, aperms
16518 Set read/write permissions for the output frames.
16520 These filters are mainly aimed at developers to test direct path in the
16521 following filter in the filtergraph.
16523 The filters accept the following options:
16527 Select the permissions mode.
16529 It accepts the following values:
16532 Do nothing. This is the default.
16534 Set all the output frames read-only.
16536 Set all the output frames directly writable.
16538 Make the frame read-only if writable, and writable if read-only.
16540 Set each output frame read-only or writable randomly.
16544 Set the seed for the @var{random} mode, must be an integer included between
16545 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16546 @code{-1}, the filter will try to use a good random seed on a best effort
16550 Note: in case of auto-inserted filter between the permission filter and the
16551 following one, the permission might not be received as expected in that
16552 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16553 perms/aperms filter can avoid this problem.
16555 @section realtime, arealtime
16557 Slow down filtering to match real time approximatively.
16559 These filters will pause the filtering for a variable amount of time to
16560 match the output rate with the input timestamps.
16561 They are similar to the @option{re} option to @code{ffmpeg}.
16563 They accept the following options:
16567 Time limit for the pauses. Any pause longer than that will be considered
16568 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16572 @section select, aselect
16574 Select frames to pass in output.
16576 This filter accepts the following options:
16581 Set expression, which is evaluated for each input frame.
16583 If the expression is evaluated to zero, the frame is discarded.
16585 If the evaluation result is negative or NaN, the frame is sent to the
16586 first output; otherwise it is sent to the output with index
16587 @code{ceil(val)-1}, assuming that the input index starts from 0.
16589 For example a value of @code{1.2} corresponds to the output with index
16590 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16593 Set the number of outputs. The output to which to send the selected
16594 frame is based on the result of the evaluation. Default value is 1.
16597 The expression can contain the following constants:
16601 The (sequential) number of the filtered frame, starting from 0.
16604 The (sequential) number of the selected frame, starting from 0.
16606 @item prev_selected_n
16607 The sequential number of the last selected frame. It's NAN if undefined.
16610 The timebase of the input timestamps.
16613 The PTS (Presentation TimeStamp) of the filtered video frame,
16614 expressed in @var{TB} units. It's NAN if undefined.
16617 The PTS of the filtered video frame,
16618 expressed in seconds. It's NAN if undefined.
16621 The PTS of the previously filtered video frame. It's NAN if undefined.
16623 @item prev_selected_pts
16624 The PTS of the last previously filtered video frame. It's NAN if undefined.
16626 @item prev_selected_t
16627 The PTS of the last previously selected video frame. It's NAN if undefined.
16630 The PTS of the first video frame in the video. It's NAN if undefined.
16633 The time of the first video frame in the video. It's NAN if undefined.
16635 @item pict_type @emph{(video only)}
16636 The type of the filtered frame. It can assume one of the following
16648 @item interlace_type @emph{(video only)}
16649 The frame interlace type. It can assume one of the following values:
16652 The frame is progressive (not interlaced).
16654 The frame is top-field-first.
16656 The frame is bottom-field-first.
16659 @item consumed_sample_n @emph{(audio only)}
16660 the number of selected samples before the current frame
16662 @item samples_n @emph{(audio only)}
16663 the number of samples in the current frame
16665 @item sample_rate @emph{(audio only)}
16666 the input sample rate
16669 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16672 the position in the file of the filtered frame, -1 if the information
16673 is not available (e.g. for synthetic video)
16675 @item scene @emph{(video only)}
16676 value between 0 and 1 to indicate a new scene; a low value reflects a low
16677 probability for the current frame to introduce a new scene, while a higher
16678 value means the current frame is more likely to be one (see the example below)
16680 @item concatdec_select
16681 The concat demuxer can select only part of a concat input file by setting an
16682 inpoint and an outpoint, but the output packets may not be entirely contained
16683 in the selected interval. By using this variable, it is possible to skip frames
16684 generated by the concat demuxer which are not exactly contained in the selected
16687 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16688 and the @var{lavf.concat.duration} packet metadata values which are also
16689 present in the decoded frames.
16691 The @var{concatdec_select} variable is -1 if the frame pts is at least
16692 start_time and either the duration metadata is missing or the frame pts is less
16693 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16696 That basically means that an input frame is selected if its pts is within the
16697 interval set by the concat demuxer.
16701 The default value of the select expression is "1".
16703 @subsection Examples
16707 Select all frames in input:
16712 The example above is the same as:
16724 Select only I-frames:
16726 select='eq(pict_type\,I)'
16730 Select one frame every 100:
16732 select='not(mod(n\,100))'
16736 Select only frames contained in the 10-20 time interval:
16738 select=between(t\,10\,20)
16742 Select only I-frames contained in the 10-20 time interval:
16744 select=between(t\,10\,20)*eq(pict_type\,I)
16748 Select frames with a minimum distance of 10 seconds:
16750 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16754 Use aselect to select only audio frames with samples number > 100:
16756 aselect='gt(samples_n\,100)'
16760 Create a mosaic of the first scenes:
16762 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16765 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16769 Send even and odd frames to separate outputs, and compose them:
16771 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16775 Select useful frames from an ffconcat file which is using inpoints and
16776 outpoints but where the source files are not intra frame only.
16778 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16782 @section sendcmd, asendcmd
16784 Send commands to filters in the filtergraph.
16786 These filters read commands to be sent to other filters in the
16789 @code{sendcmd} must be inserted between two video filters,
16790 @code{asendcmd} must be inserted between two audio filters, but apart
16791 from that they act the same way.
16793 The specification of commands can be provided in the filter arguments
16794 with the @var{commands} option, or in a file specified by the
16795 @var{filename} option.
16797 These filters accept the following options:
16800 Set the commands to be read and sent to the other filters.
16802 Set the filename of the commands to be read and sent to the other
16806 @subsection Commands syntax
16808 A commands description consists of a sequence of interval
16809 specifications, comprising a list of commands to be executed when a
16810 particular event related to that interval occurs. The occurring event
16811 is typically the current frame time entering or leaving a given time
16814 An interval is specified by the following syntax:
16816 @var{START}[-@var{END}] @var{COMMANDS};
16819 The time interval is specified by the @var{START} and @var{END} times.
16820 @var{END} is optional and defaults to the maximum time.
16822 The current frame time is considered within the specified interval if
16823 it is included in the interval [@var{START}, @var{END}), that is when
16824 the time is greater or equal to @var{START} and is lesser than
16827 @var{COMMANDS} consists of a sequence of one or more command
16828 specifications, separated by ",", relating to that interval. The
16829 syntax of a command specification is given by:
16831 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16834 @var{FLAGS} is optional and specifies the type of events relating to
16835 the time interval which enable sending the specified command, and must
16836 be a non-null sequence of identifier flags separated by "+" or "|" and
16837 enclosed between "[" and "]".
16839 The following flags are recognized:
16842 The command is sent when the current frame timestamp enters the
16843 specified interval. In other words, the command is sent when the
16844 previous frame timestamp was not in the given interval, and the
16848 The command is sent when the current frame timestamp leaves the
16849 specified interval. In other words, the command is sent when the
16850 previous frame timestamp was in the given interval, and the
16854 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16857 @var{TARGET} specifies the target of the command, usually the name of
16858 the filter class or a specific filter instance name.
16860 @var{COMMAND} specifies the name of the command for the target filter.
16862 @var{ARG} is optional and specifies the optional list of argument for
16863 the given @var{COMMAND}.
16865 Between one interval specification and another, whitespaces, or
16866 sequences of characters starting with @code{#} until the end of line,
16867 are ignored and can be used to annotate comments.
16869 A simplified BNF description of the commands specification syntax
16872 @var{COMMAND_FLAG} ::= "enter" | "leave"
16873 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16874 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16875 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16876 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16877 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16880 @subsection Examples
16884 Specify audio tempo change at second 4:
16886 asendcmd=c='4.0 atempo tempo 1.5',atempo
16890 Specify a list of drawtext and hue commands in a file.
16892 # show text in the interval 5-10
16893 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16894 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16896 # desaturate the image in the interval 15-20
16897 15.0-20.0 [enter] hue s 0,
16898 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16900 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16902 # apply an exponential saturation fade-out effect, starting from time 25
16903 25 [enter] hue s exp(25-t)
16906 A filtergraph allowing to read and process the above command list
16907 stored in a file @file{test.cmd}, can be specified with:
16909 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
16914 @section setpts, asetpts
16916 Change the PTS (presentation timestamp) of the input frames.
16918 @code{setpts} works on video frames, @code{asetpts} on audio frames.
16920 This filter accepts the following options:
16925 The expression which is evaluated for each frame to construct its timestamp.
16929 The expression is evaluated through the eval API and can contain the following
16934 frame rate, only defined for constant frame-rate video
16937 The presentation timestamp in input
16940 The count of the input frame for video or the number of consumed samples,
16941 not including the current frame for audio, starting from 0.
16943 @item NB_CONSUMED_SAMPLES
16944 The number of consumed samples, not including the current frame (only
16947 @item NB_SAMPLES, S
16948 The number of samples in the current frame (only audio)
16950 @item SAMPLE_RATE, SR
16951 The audio sample rate.
16954 The PTS of the first frame.
16957 the time in seconds of the first frame
16960 State whether the current frame is interlaced.
16963 the time in seconds of the current frame
16966 original position in the file of the frame, or undefined if undefined
16967 for the current frame
16970 The previous input PTS.
16973 previous input time in seconds
16976 The previous output PTS.
16979 previous output time in seconds
16982 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
16986 The wallclock (RTC) time at the start of the movie in microseconds.
16989 The timebase of the input timestamps.
16993 @subsection Examples
16997 Start counting PTS from zero
16999 setpts=PTS-STARTPTS
17003 Apply fast motion effect:
17009 Apply slow motion effect:
17015 Set fixed rate of 25 frames per second:
17021 Set fixed rate 25 fps with some jitter:
17023 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17027 Apply an offset of 10 seconds to the input PTS:
17033 Generate timestamps from a "live source" and rebase onto the current timebase:
17035 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17039 Generate timestamps by counting samples:
17046 @section settb, asettb
17048 Set the timebase to use for the output frames timestamps.
17049 It is mainly useful for testing timebase configuration.
17051 It accepts the following parameters:
17056 The expression which is evaluated into the output timebase.
17060 The value for @option{tb} is an arithmetic expression representing a
17061 rational. The expression can contain the constants "AVTB" (the default
17062 timebase), "intb" (the input timebase) and "sr" (the sample rate,
17063 audio only). Default value is "intb".
17065 @subsection Examples
17069 Set the timebase to 1/25:
17075 Set the timebase to 1/10:
17081 Set the timebase to 1001/1000:
17087 Set the timebase to 2*intb:
17093 Set the default timebase value:
17100 Convert input audio to a video output representing frequency spectrum
17101 logarithmically using Brown-Puckette constant Q transform algorithm with
17102 direct frequency domain coefficient calculation (but the transform itself
17103 is not really constant Q, instead the Q factor is actually variable/clamped),
17104 with musical tone scale, from E0 to D#10.
17106 The filter accepts the following options:
17110 Specify the video size for the output. It must be even. For the syntax of this option,
17111 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17112 Default value is @code{1920x1080}.
17115 Set the output frame rate. Default value is @code{25}.
17118 Set the bargraph height. It must be even. Default value is @code{-1} which
17119 computes the bargraph height automatically.
17122 Set the axis height. It must be even. Default value is @code{-1} which computes
17123 the axis height automatically.
17126 Set the sonogram height. It must be even. Default value is @code{-1} which
17127 computes the sonogram height automatically.
17130 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17131 instead. Default value is @code{1}.
17133 @item sono_v, volume
17134 Specify the sonogram volume expression. It can contain variables:
17137 the @var{bar_v} evaluated expression
17138 @item frequency, freq, f
17139 the frequency where it is evaluated
17140 @item timeclamp, tc
17141 the value of @var{timeclamp} option
17145 @item a_weighting(f)
17146 A-weighting of equal loudness
17147 @item b_weighting(f)
17148 B-weighting of equal loudness
17149 @item c_weighting(f)
17150 C-weighting of equal loudness.
17152 Default value is @code{16}.
17154 @item bar_v, volume2
17155 Specify the bargraph volume expression. It can contain variables:
17158 the @var{sono_v} evaluated expression
17159 @item frequency, freq, f
17160 the frequency where it is evaluated
17161 @item timeclamp, tc
17162 the value of @var{timeclamp} option
17166 @item a_weighting(f)
17167 A-weighting of equal loudness
17168 @item b_weighting(f)
17169 B-weighting of equal loudness
17170 @item c_weighting(f)
17171 C-weighting of equal loudness.
17173 Default value is @code{sono_v}.
17175 @item sono_g, gamma
17176 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17177 higher gamma makes the spectrum having more range. Default value is @code{3}.
17178 Acceptable range is @code{[1, 7]}.
17180 @item bar_g, gamma2
17181 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17185 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17186 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17188 @item timeclamp, tc
17189 Specify the transform timeclamp. At low frequency, there is trade-off between
17190 accuracy in time domain and frequency domain. If timeclamp is lower,
17191 event in time domain is represented more accurately (such as fast bass drum),
17192 otherwise event in frequency domain is represented more accurately
17193 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17196 Set attack time in seconds. The default is @code{0} (disabled). Otherwise, it
17197 limits future samples by applying asymmetric windowing in time domain, useful
17198 when low latency is required. Accepted range is @code{[0, 1]}.
17201 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17202 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17205 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17206 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17209 This option is deprecated and ignored.
17212 Specify the transform length in time domain. Use this option to control accuracy
17213 trade-off between time domain and frequency domain at every frequency sample.
17214 It can contain variables:
17216 @item frequency, freq, f
17217 the frequency where it is evaluated
17218 @item timeclamp, tc
17219 the value of @var{timeclamp} option.
17221 Default value is @code{384*tc/(384+tc*f)}.
17224 Specify the transform count for every video frame. Default value is @code{6}.
17225 Acceptable range is @code{[1, 30]}.
17228 Specify the transform count for every single pixel. Default value is @code{0},
17229 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17232 Specify font file for use with freetype to draw the axis. If not specified,
17233 use embedded font. Note that drawing with font file or embedded font is not
17234 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17238 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17239 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17242 Specify font color expression. This is arithmetic expression that should return
17243 integer value 0xRRGGBB. It can contain variables:
17245 @item frequency, freq, f
17246 the frequency where it is evaluated
17247 @item timeclamp, tc
17248 the value of @var{timeclamp} option
17253 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17254 @item r(x), g(x), b(x)
17255 red, green, and blue value of intensity x.
17257 Default value is @code{st(0, (midi(f)-59.5)/12);
17258 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17259 r(1-ld(1)) + b(ld(1))}.
17262 Specify image file to draw the axis. This option override @var{fontfile} and
17263 @var{fontcolor} option.
17266 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17267 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17268 Default value is @code{1}.
17271 Set colorspace. The accepted values are:
17274 Unspecified (default)
17283 BT.470BG or BT.601-6 625
17286 SMPTE-170M or BT.601-6 525
17292 BT.2020 with non-constant luminance
17297 Set spectrogram color scheme. This is list of floating point values with format
17298 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17299 The default is @code{1|0.5|0|0|0.5|1}.
17303 @subsection Examples
17307 Playing audio while showing the spectrum:
17309 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17313 Same as above, but with frame rate 30 fps:
17315 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17319 Playing at 1280x720:
17321 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17325 Disable sonogram display:
17331 A1 and its harmonics: A1, A2, (near)E3, A3:
17333 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),
17334 asplit[a][out1]; [a] showcqt [out0]'
17338 Same as above, but with more accuracy in frequency domain:
17340 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),
17341 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17347 bar_v=10:sono_v=bar_v*a_weighting(f)
17351 Custom gamma, now spectrum is linear to the amplitude.
17357 Custom tlength equation:
17359 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)))'
17363 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17365 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17369 Custom font using fontconfig:
17371 font='Courier New,Monospace,mono|bold'
17375 Custom frequency range with custom axis using image file:
17377 axisfile=myaxis.png:basefreq=40:endfreq=10000
17383 Convert input audio to video output representing the audio power spectrum.
17384 Audio amplitude is on Y-axis while frequency is on X-axis.
17386 The filter accepts the following options:
17390 Specify size of video. For the syntax of this option, check the
17391 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17392 Default is @code{1024x512}.
17396 This set how each frequency bin will be represented.
17398 It accepts the following values:
17404 Default is @code{bar}.
17407 Set amplitude scale.
17409 It accepts the following values:
17423 Default is @code{log}.
17426 Set frequency scale.
17428 It accepts the following values:
17437 Reverse logarithmic scale.
17439 Default is @code{lin}.
17444 It accepts the following values:
17460 Default is @code{w2048}
17463 Set windowing function.
17465 It accepts the following values:
17487 Default is @code{hanning}.
17490 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17491 which means optimal overlap for selected window function will be picked.
17494 Set time averaging. Setting this to 0 will display current maximal peaks.
17495 Default is @code{1}, which means time averaging is disabled.
17498 Specify list of colors separated by space or by '|' which will be used to
17499 draw channel frequencies. Unrecognized or missing colors will be replaced
17503 Set channel display mode.
17505 It accepts the following values:
17510 Default is @code{combined}.
17513 Set minimum amplitude used in @code{log} amplitude scaler.
17517 @anchor{showspectrum}
17518 @section showspectrum
17520 Convert input audio to a video output, representing the audio frequency
17523 The filter accepts the following options:
17527 Specify the video size for the output. For the syntax of this option, check the
17528 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17529 Default value is @code{640x512}.
17532 Specify how the spectrum should slide along the window.
17534 It accepts the following values:
17537 the samples start again on the left when they reach the right
17539 the samples scroll from right to left
17541 frames are only produced when the samples reach the right
17543 the samples scroll from left to right
17546 Default value is @code{replace}.
17549 Specify display mode.
17551 It accepts the following values:
17554 all channels are displayed in the same row
17556 all channels are displayed in separate rows
17559 Default value is @samp{combined}.
17562 Specify display color mode.
17564 It accepts the following values:
17567 each channel is displayed in a separate color
17569 each channel is displayed using the same color scheme
17571 each channel is displayed using the rainbow color scheme
17573 each channel is displayed using the moreland color scheme
17575 each channel is displayed using the nebulae color scheme
17577 each channel is displayed using the fire color scheme
17579 each channel is displayed using the fiery color scheme
17581 each channel is displayed using the fruit color scheme
17583 each channel is displayed using the cool color scheme
17586 Default value is @samp{channel}.
17589 Specify scale used for calculating intensity color values.
17591 It accepts the following values:
17596 square root, default
17607 Default value is @samp{sqrt}.
17610 Set saturation modifier for displayed colors. Negative values provide
17611 alternative color scheme. @code{0} is no saturation at all.
17612 Saturation must be in [-10.0, 10.0] range.
17613 Default value is @code{1}.
17616 Set window function.
17618 It accepts the following values:
17642 Default value is @code{hann}.
17645 Set orientation of time vs frequency axis. Can be @code{vertical} or
17646 @code{horizontal}. Default is @code{vertical}.
17649 Set ratio of overlap window. Default value is @code{0}.
17650 When value is @code{1} overlap is set to recommended size for specific
17651 window function currently used.
17654 Set scale gain for calculating intensity color values.
17655 Default value is @code{1}.
17658 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17661 Set color rotation, must be in [-1.0, 1.0] range.
17662 Default value is @code{0}.
17665 The usage is very similar to the showwaves filter; see the examples in that
17668 @subsection Examples
17672 Large window with logarithmic color scaling:
17674 showspectrum=s=1280x480:scale=log
17678 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17680 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17681 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17685 @section showspectrumpic
17687 Convert input audio to a single video frame, representing the audio frequency
17690 The filter accepts the following options:
17694 Specify the video size for the output. For the syntax of this option, check the
17695 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17696 Default value is @code{4096x2048}.
17699 Specify display mode.
17701 It accepts the following values:
17704 all channels are displayed in the same row
17706 all channels are displayed in separate rows
17708 Default value is @samp{combined}.
17711 Specify display color mode.
17713 It accepts the following values:
17716 each channel is displayed in a separate color
17718 each channel is displayed using the same color scheme
17720 each channel is displayed using the rainbow color scheme
17722 each channel is displayed using the moreland color scheme
17724 each channel is displayed using the nebulae color scheme
17726 each channel is displayed using the fire color scheme
17728 each channel is displayed using the fiery color scheme
17730 each channel is displayed using the fruit color scheme
17732 each channel is displayed using the cool color scheme
17734 Default value is @samp{intensity}.
17737 Specify scale used for calculating intensity color values.
17739 It accepts the following values:
17744 square root, default
17754 Default value is @samp{log}.
17757 Set saturation modifier for displayed colors. Negative values provide
17758 alternative color scheme. @code{0} is no saturation at all.
17759 Saturation must be in [-10.0, 10.0] range.
17760 Default value is @code{1}.
17763 Set window function.
17765 It accepts the following values:
17788 Default value is @code{hann}.
17791 Set orientation of time vs frequency axis. Can be @code{vertical} or
17792 @code{horizontal}. Default is @code{vertical}.
17795 Set scale gain for calculating intensity color values.
17796 Default value is @code{1}.
17799 Draw time and frequency axes and legends. Default is enabled.
17802 Set color rotation, must be in [-1.0, 1.0] range.
17803 Default value is @code{0}.
17806 @subsection Examples
17810 Extract an audio spectrogram of a whole audio track
17811 in a 1024x1024 picture using @command{ffmpeg}:
17813 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17817 @section showvolume
17819 Convert input audio volume to a video output.
17821 The filter accepts the following options:
17828 Set border width, allowed range is [0, 5]. Default is 1.
17831 Set channel width, allowed range is [80, 8192]. Default is 400.
17834 Set channel height, allowed range is [1, 900]. Default is 20.
17837 Set fade, allowed range is [0.001, 1]. Default is 0.95.
17840 Set volume color expression.
17842 The expression can use the following variables:
17846 Current max volume of channel in dB.
17852 Current channel number, starting from 0.
17856 If set, displays channel names. Default is enabled.
17859 If set, displays volume values. Default is enabled.
17862 Set orientation, can be @code{horizontal} or @code{vertical},
17863 default is @code{horizontal}.
17866 Set step size, allowed range s [0, 5]. Default is 0, which means
17872 Convert input audio to a video output, representing the samples waves.
17874 The filter accepts the following options:
17878 Specify the video size for the output. For the syntax of this option, check the
17879 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17880 Default value is @code{600x240}.
17885 Available values are:
17888 Draw a point for each sample.
17891 Draw a vertical line for each sample.
17894 Draw a point for each sample and a line between them.
17897 Draw a centered vertical line for each sample.
17900 Default value is @code{point}.
17903 Set the number of samples which are printed on the same column. A
17904 larger value will decrease the frame rate. Must be a positive
17905 integer. This option can be set only if the value for @var{rate}
17906 is not explicitly specified.
17909 Set the (approximate) output frame rate. This is done by setting the
17910 option @var{n}. Default value is "25".
17912 @item split_channels
17913 Set if channels should be drawn separately or overlap. Default value is 0.
17916 Set colors separated by '|' which are going to be used for drawing of each channel.
17919 Set amplitude scale.
17921 Available values are:
17939 @subsection Examples
17943 Output the input file audio and the corresponding video representation
17946 amovie=a.mp3,asplit[out0],showwaves[out1]
17950 Create a synthetic signal and show it with showwaves, forcing a
17951 frame rate of 30 frames per second:
17953 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
17957 @section showwavespic
17959 Convert input audio to a single video frame, representing the samples waves.
17961 The filter accepts the following options:
17965 Specify the video size for the output. For the syntax of this option, check the
17966 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17967 Default value is @code{600x240}.
17969 @item split_channels
17970 Set if channels should be drawn separately or overlap. Default value is 0.
17973 Set colors separated by '|' which are going to be used for drawing of each channel.
17976 Set amplitude scale.
17978 Available values are:
17996 @subsection Examples
18000 Extract a channel split representation of the wave form of a whole audio track
18001 in a 1024x800 picture using @command{ffmpeg}:
18003 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
18007 @section sidedata, asidedata
18009 Delete frame side data, or select frames based on it.
18011 This filter accepts the following options:
18015 Set mode of operation of the filter.
18017 Can be one of the following:
18021 Select every frame with side data of @code{type}.
18024 Delete side data of @code{type}. If @code{type} is not set, delete all side
18030 Set side data type used with all modes. Must be set for @code{select} mode. For
18031 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18032 in @file{libavutil/frame.h}. For example, to choose
18033 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18037 @section spectrumsynth
18039 Sythesize audio from 2 input video spectrums, first input stream represents
18040 magnitude across time and second represents phase across time.
18041 The filter will transform from frequency domain as displayed in videos back
18042 to time domain as presented in audio output.
18044 This filter is primarily created for reversing processed @ref{showspectrum}
18045 filter outputs, but can synthesize sound from other spectrograms too.
18046 But in such case results are going to be poor if the phase data is not
18047 available, because in such cases phase data need to be recreated, usually
18048 its just recreated from random noise.
18049 For best results use gray only output (@code{channel} color mode in
18050 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
18051 @code{lin} scale for phase video. To produce phase, for 2nd video, use
18052 @code{data} option. Inputs videos should generally use @code{fullframe}
18053 slide mode as that saves resources needed for decoding video.
18055 The filter accepts the following options:
18059 Specify sample rate of output audio, the sample rate of audio from which
18060 spectrum was generated may differ.
18063 Set number of channels represented in input video spectrums.
18066 Set scale which was used when generating magnitude input spectrum.
18067 Can be @code{lin} or @code{log}. Default is @code{log}.
18070 Set slide which was used when generating inputs spectrums.
18071 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18072 Default is @code{fullframe}.
18075 Set window function used for resynthesis.
18078 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18079 which means optimal overlap for selected window function will be picked.
18082 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18083 Default is @code{vertical}.
18086 @subsection Examples
18090 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18091 then resynthesize videos back to audio with spectrumsynth:
18093 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
18094 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
18095 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18099 @section split, asplit
18101 Split input into several identical outputs.
18103 @code{asplit} works with audio input, @code{split} with video.
18105 The filter accepts a single parameter which specifies the number of outputs. If
18106 unspecified, it defaults to 2.
18108 @subsection Examples
18112 Create two separate outputs from the same input:
18114 [in] split [out0][out1]
18118 To create 3 or more outputs, you need to specify the number of
18121 [in] asplit=3 [out0][out1][out2]
18125 Create two separate outputs from the same input, one cropped and
18128 [in] split [splitout1][splitout2];
18129 [splitout1] crop=100:100:0:0 [cropout];
18130 [splitout2] pad=200:200:100:100 [padout];
18134 Create 5 copies of the input audio with @command{ffmpeg}:
18136 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18142 Receive commands sent through a libzmq client, and forward them to
18143 filters in the filtergraph.
18145 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18146 must be inserted between two video filters, @code{azmq} between two
18149 To enable these filters you need to install the libzmq library and
18150 headers and configure FFmpeg with @code{--enable-libzmq}.
18152 For more information about libzmq see:
18153 @url{http://www.zeromq.org/}
18155 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18156 receives messages sent through a network interface defined by the
18157 @option{bind_address} option.
18159 The received message must be in the form:
18161 @var{TARGET} @var{COMMAND} [@var{ARG}]
18164 @var{TARGET} specifies the target of the command, usually the name of
18165 the filter class or a specific filter instance name.
18167 @var{COMMAND} specifies the name of the command for the target filter.
18169 @var{ARG} is optional and specifies the optional argument list for the
18170 given @var{COMMAND}.
18172 Upon reception, the message is processed and the corresponding command
18173 is injected into the filtergraph. Depending on the result, the filter
18174 will send a reply to the client, adopting the format:
18176 @var{ERROR_CODE} @var{ERROR_REASON}
18180 @var{MESSAGE} is optional.
18182 @subsection Examples
18184 Look at @file{tools/zmqsend} for an example of a zmq client which can
18185 be used to send commands processed by these filters.
18187 Consider the following filtergraph generated by @command{ffplay}
18189 ffplay -dumpgraph 1 -f lavfi "
18190 color=s=100x100:c=red [l];
18191 color=s=100x100:c=blue [r];
18192 nullsrc=s=200x100, zmq [bg];
18193 [bg][l] overlay [bg+l];
18194 [bg+l][r] overlay=x=100 "
18197 To change the color of the left side of the video, the following
18198 command can be used:
18200 echo Parsed_color_0 c yellow | tools/zmqsend
18203 To change the right side:
18205 echo Parsed_color_1 c pink | tools/zmqsend
18208 @c man end MULTIMEDIA FILTERS
18210 @chapter Multimedia Sources
18211 @c man begin MULTIMEDIA SOURCES
18213 Below is a description of the currently available multimedia sources.
18217 This is the same as @ref{movie} source, except it selects an audio
18223 Read audio and/or video stream(s) from a movie container.
18225 It accepts the following parameters:
18229 The name of the resource to read (not necessarily a file; it can also be a
18230 device or a stream accessed through some protocol).
18232 @item format_name, f
18233 Specifies the format assumed for the movie to read, and can be either
18234 the name of a container or an input device. If not specified, the
18235 format is guessed from @var{movie_name} or by probing.
18237 @item seek_point, sp
18238 Specifies the seek point in seconds. The frames will be output
18239 starting from this seek point. The parameter is evaluated with
18240 @code{av_strtod}, so the numerical value may be suffixed by an IS
18241 postfix. The default value is "0".
18244 Specifies the streams to read. Several streams can be specified,
18245 separated by "+". The source will then have as many outputs, in the
18246 same order. The syntax is explained in the ``Stream specifiers''
18247 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18248 respectively the default (best suited) video and audio stream. Default
18249 is "dv", or "da" if the filter is called as "amovie".
18251 @item stream_index, si
18252 Specifies the index of the video stream to read. If the value is -1,
18253 the most suitable video stream will be automatically selected. The default
18254 value is "-1". Deprecated. If the filter is called "amovie", it will select
18255 audio instead of video.
18258 Specifies how many times to read the stream in sequence.
18259 If the value is 0, the stream will be looped infinitely.
18260 Default value is "1".
18262 Note that when the movie is looped the source timestamps are not
18263 changed, so it will generate non monotonically increasing timestamps.
18265 @item discontinuity
18266 Specifies the time difference between frames above which the point is
18267 considered a timestamp discontinuity which is removed by adjusting the later
18271 It allows overlaying a second video on top of the main input of
18272 a filtergraph, as shown in this graph:
18274 input -----------> deltapts0 --> overlay --> output
18277 movie --> scale--> deltapts1 -------+
18279 @subsection Examples
18283 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18284 on top of the input labelled "in":
18286 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18287 [in] setpts=PTS-STARTPTS [main];
18288 [main][over] overlay=16:16 [out]
18292 Read from a video4linux2 device, and overlay it on top of the input
18295 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18296 [in] setpts=PTS-STARTPTS [main];
18297 [main][over] overlay=16:16 [out]
18301 Read the first video stream and the audio stream with id 0x81 from
18302 dvd.vob; the video is connected to the pad named "video" and the audio is
18303 connected to the pad named "audio":
18305 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18309 @subsection Commands
18311 Both movie and amovie support the following commands:
18314 Perform seek using "av_seek_frame".
18315 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18318 @var{stream_index}: If stream_index is -1, a default
18319 stream is selected, and @var{timestamp} is automatically converted
18320 from AV_TIME_BASE units to the stream specific time_base.
18322 @var{timestamp}: Timestamp in AVStream.time_base units
18323 or, if no stream is specified, in AV_TIME_BASE units.
18325 @var{flags}: Flags which select direction and seeking mode.
18329 Get movie duration in AV_TIME_BASE units.
18333 @c man end MULTIMEDIA SOURCES