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.
10434 Specify the color of the padded area. For the syntax of this option,
10435 check the "Color" section in the ffmpeg-utils manual.
10437 The default value of @var{color} is "black".
10440 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10442 It accepts the following values:
10446 Only evaluate expressions once during the filter initialization or when
10447 a command is processed.
10450 Evaluate expressions for each incoming frame.
10454 Default value is @samp{init}.
10458 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10459 options are expressions containing the following constants:
10464 The input video width and height.
10468 These are the same as @var{in_w} and @var{in_h}.
10472 The output width and height (the size of the padded area), as
10473 specified by the @var{width} and @var{height} expressions.
10477 These are the same as @var{out_w} and @var{out_h}.
10481 The x and y offsets as specified by the @var{x} and @var{y}
10482 expressions, or NAN if not yet specified.
10485 same as @var{iw} / @var{ih}
10488 input sample aspect ratio
10491 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10495 The horizontal and vertical chroma subsample values. For example for the
10496 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10499 @subsection Examples
10503 Add paddings with the color "violet" to the input video. The output video
10504 size is 640x480, and the top-left corner of the input video is placed at
10507 pad=640:480:0:40:violet
10510 The example above is equivalent to the following command:
10512 pad=width=640:height=480:x=0:y=40:color=violet
10516 Pad the input to get an output with dimensions increased by 3/2,
10517 and put the input video at the center of the padded area:
10519 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10523 Pad the input to get a squared output with size equal to the maximum
10524 value between the input width and height, and put the input video at
10525 the center of the padded area:
10527 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10531 Pad the input to get a final w/h ratio of 16:9:
10533 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10537 In case of anamorphic video, in order to set the output display aspect
10538 correctly, it is necessary to use @var{sar} in the expression,
10539 according to the relation:
10541 (ih * X / ih) * sar = output_dar
10542 X = output_dar / sar
10545 Thus the previous example needs to be modified to:
10547 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10551 Double the output size and put the input video in the bottom-right
10552 corner of the output padded area:
10554 pad="2*iw:2*ih:ow-iw:oh-ih"
10558 @anchor{palettegen}
10559 @section palettegen
10561 Generate one palette for a whole video stream.
10563 It accepts the following options:
10567 Set the maximum number of colors to quantize in the palette.
10568 Note: the palette will still contain 256 colors; the unused palette entries
10571 @item reserve_transparent
10572 Create a palette of 255 colors maximum and reserve the last one for
10573 transparency. Reserving the transparency color is useful for GIF optimization.
10574 If not set, the maximum of colors in the palette will be 256. You probably want
10575 to disable this option for a standalone image.
10579 Set statistics mode.
10581 It accepts the following values:
10584 Compute full frame histograms.
10586 Compute histograms only for the part that differs from previous frame. This
10587 might be relevant to give more importance to the moving part of your input if
10588 the background is static.
10590 Compute new histogram for each frame.
10593 Default value is @var{full}.
10596 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10597 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10598 color quantization of the palette. This information is also visible at
10599 @var{info} logging level.
10601 @subsection Examples
10605 Generate a representative palette of a given video using @command{ffmpeg}:
10607 ffmpeg -i input.mkv -vf palettegen palette.png
10611 @section paletteuse
10613 Use a palette to downsample an input video stream.
10615 The filter takes two inputs: one video stream and a palette. The palette must
10616 be a 256 pixels image.
10618 It accepts the following options:
10622 Select dithering mode. Available algorithms are:
10625 Ordered 8x8 bayer dithering (deterministic)
10627 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10628 Note: this dithering is sometimes considered "wrong" and is included as a
10630 @item floyd_steinberg
10631 Floyd and Steingberg dithering (error diffusion)
10633 Frankie Sierra dithering v2 (error diffusion)
10635 Frankie Sierra dithering v2 "Lite" (error diffusion)
10638 Default is @var{sierra2_4a}.
10641 When @var{bayer} dithering is selected, this option defines the scale of the
10642 pattern (how much the crosshatch pattern is visible). A low value means more
10643 visible pattern for less banding, and higher value means less visible pattern
10644 at the cost of more banding.
10646 The option must be an integer value in the range [0,5]. Default is @var{2}.
10649 If set, define the zone to process
10653 Only the changing rectangle will be reprocessed. This is similar to GIF
10654 cropping/offsetting compression mechanism. This option can be useful for speed
10655 if only a part of the image is changing, and has use cases such as limiting the
10656 scope of the error diffusal @option{dither} to the rectangle that bounds the
10657 moving scene (it leads to more deterministic output if the scene doesn't change
10658 much, and as a result less moving noise and better GIF compression).
10661 Default is @var{none}.
10664 Take new palette for each output frame.
10667 @subsection Examples
10671 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10672 using @command{ffmpeg}:
10674 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10678 @section perspective
10680 Correct perspective of video not recorded perpendicular to the screen.
10682 A description of the accepted parameters follows.
10693 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10694 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10695 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10696 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10697 then the corners of the source will be sent to the specified coordinates.
10699 The expressions can use the following variables:
10704 the width and height of video frame.
10708 Output frame count.
10711 @item interpolation
10712 Set interpolation for perspective correction.
10714 It accepts the following values:
10720 Default value is @samp{linear}.
10723 Set interpretation of coordinate options.
10725 It accepts the following values:
10729 Send point in the source specified by the given coordinates to
10730 the corners of the destination.
10732 @item 1, destination
10734 Send the corners of the source to the point in the destination specified
10735 by the given coordinates.
10737 Default value is @samp{source}.
10741 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10743 It accepts the following values:
10746 only evaluate expressions once during the filter initialization or
10747 when a command is processed
10750 evaluate expressions for each incoming frame
10753 Default value is @samp{init}.
10758 Delay interlaced video by one field time so that the field order changes.
10760 The intended use is to fix PAL movies that have been captured with the
10761 opposite field order to the film-to-video transfer.
10763 A description of the accepted parameters follows.
10769 It accepts the following values:
10772 Capture field order top-first, transfer bottom-first.
10773 Filter will delay the bottom field.
10776 Capture field order bottom-first, transfer top-first.
10777 Filter will delay the top field.
10780 Capture and transfer with the same field order. This mode only exists
10781 for the documentation of the other options to refer to, but if you
10782 actually select it, the filter will faithfully do nothing.
10785 Capture field order determined automatically by field flags, transfer
10787 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10788 basis using field flags. If no field information is available,
10789 then this works just like @samp{u}.
10792 Capture unknown or varying, transfer opposite.
10793 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10794 analyzing the images and selecting the alternative that produces best
10795 match between the fields.
10798 Capture top-first, transfer unknown or varying.
10799 Filter selects among @samp{t} and @samp{p} using image analysis.
10802 Capture bottom-first, transfer unknown or varying.
10803 Filter selects among @samp{b} and @samp{p} using image analysis.
10806 Capture determined by field flags, transfer unknown or varying.
10807 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10808 image analysis. If no field information is available, then this works just
10809 like @samp{U}. This is the default mode.
10812 Both capture and transfer unknown or varying.
10813 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10817 @section pixdesctest
10819 Pixel format descriptor test filter, mainly useful for internal
10820 testing. The output video should be equal to the input video.
10824 format=monow, pixdesctest
10827 can be used to test the monowhite pixel format descriptor definition.
10831 Enable the specified chain of postprocessing subfilters using libpostproc. This
10832 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10833 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10834 Each subfilter and some options have a short and a long name that can be used
10835 interchangeably, i.e. dr/dering are the same.
10837 The filters accept the following options:
10841 Set postprocessing subfilters string.
10844 All subfilters share common options to determine their scope:
10848 Honor the quality commands for this subfilter.
10851 Do chrominance filtering, too (default).
10854 Do luminance filtering only (no chrominance).
10857 Do chrominance filtering only (no luminance).
10860 These options can be appended after the subfilter name, separated by a '|'.
10862 Available subfilters are:
10865 @item hb/hdeblock[|difference[|flatness]]
10866 Horizontal deblocking filter
10869 Difference factor where higher values mean more deblocking (default: @code{32}).
10871 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10874 @item vb/vdeblock[|difference[|flatness]]
10875 Vertical deblocking filter
10878 Difference factor where higher values mean more deblocking (default: @code{32}).
10880 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10883 @item ha/hadeblock[|difference[|flatness]]
10884 Accurate horizontal deblocking filter
10887 Difference factor where higher values mean more deblocking (default: @code{32}).
10889 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10892 @item va/vadeblock[|difference[|flatness]]
10893 Accurate vertical deblocking filter
10896 Difference factor where higher values mean more deblocking (default: @code{32}).
10898 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10902 The horizontal and vertical deblocking filters share the difference and
10903 flatness values so you cannot set different horizontal and vertical
10907 @item h1/x1hdeblock
10908 Experimental horizontal deblocking filter
10910 @item v1/x1vdeblock
10911 Experimental vertical deblocking filter
10916 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10919 larger -> stronger filtering
10921 larger -> stronger filtering
10923 larger -> stronger filtering
10926 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10929 Stretch luminance to @code{0-255}.
10932 @item lb/linblenddeint
10933 Linear blend deinterlacing filter that deinterlaces the given block by
10934 filtering all lines with a @code{(1 2 1)} filter.
10936 @item li/linipoldeint
10937 Linear interpolating deinterlacing filter that deinterlaces the given block by
10938 linearly interpolating every second line.
10940 @item ci/cubicipoldeint
10941 Cubic interpolating deinterlacing filter deinterlaces the given block by
10942 cubically interpolating every second line.
10944 @item md/mediandeint
10945 Median deinterlacing filter that deinterlaces the given block by applying a
10946 median filter to every second line.
10948 @item fd/ffmpegdeint
10949 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10950 second line with a @code{(-1 4 2 4 -1)} filter.
10953 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10954 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10956 @item fq/forceQuant[|quantizer]
10957 Overrides the quantizer table from the input with the constant quantizer you
10965 Default pp filter combination (@code{hb|a,vb|a,dr|a})
10968 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10971 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10974 @subsection Examples
10978 Apply horizontal and vertical deblocking, deringing and automatic
10979 brightness/contrast:
10985 Apply default filters without brightness/contrast correction:
10991 Apply default filters and temporal denoiser:
10993 pp=default/tmpnoise|1|2|3
10997 Apply deblocking on luminance only, and switch vertical deblocking on or off
10998 automatically depending on available CPU time:
11005 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11006 similar to spp = 6 with 7 point DCT, where only the center sample is
11009 The filter accepts the following options:
11013 Force a constant quantization parameter. It accepts an integer in range
11014 0 to 63. If not set, the filter will use the QP from the video stream
11018 Set thresholding mode. Available modes are:
11022 Set hard thresholding.
11024 Set soft thresholding (better de-ringing effect, but likely blurrier).
11026 Set medium thresholding (good results, default).
11030 @section premultiply
11031 Apply alpha premultiply effect to input video stream using first plane
11032 of second stream as alpha.
11034 Both streams must have same dimensions and same pixel format.
11037 Apply prewitt operator to input video stream.
11039 The filter accepts the following option:
11043 Set which planes will be processed, unprocessed planes will be copied.
11044 By default value 0xf, all planes will be processed.
11047 Set value which will be multiplied with filtered result.
11050 Set value which will be added to filtered result.
11055 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11056 Ratio) between two input videos.
11058 This filter takes in input two input videos, the first input is
11059 considered the "main" source and is passed unchanged to the
11060 output. The second input is used as a "reference" video for computing
11063 Both video inputs must have the same resolution and pixel format for
11064 this filter to work correctly. Also it assumes that both inputs
11065 have the same number of frames, which are compared one by one.
11067 The obtained average PSNR is printed through the logging system.
11069 The filter stores the accumulated MSE (mean squared error) of each
11070 frame, and at the end of the processing it is averaged across all frames
11071 equally, and the following formula is applied to obtain the PSNR:
11074 PSNR = 10*log10(MAX^2/MSE)
11077 Where MAX is the average of the maximum values of each component of the
11080 The description of the accepted parameters follows.
11083 @item stats_file, f
11084 If specified the filter will use the named file to save the PSNR of
11085 each individual frame. When filename equals "-" the data is sent to
11088 @item stats_version
11089 Specifies which version of the stats file format to use. Details of
11090 each format are written below.
11091 Default value is 1.
11093 @item stats_add_max
11094 Determines whether the max value is output to the stats log.
11095 Default value is 0.
11096 Requires stats_version >= 2. If this is set and stats_version < 2,
11097 the filter will return an error.
11100 The file printed if @var{stats_file} is selected, contains a sequence of
11101 key/value pairs of the form @var{key}:@var{value} for each compared
11104 If a @var{stats_version} greater than 1 is specified, a header line precedes
11105 the list of per-frame-pair stats, with key value pairs following the frame
11106 format with the following parameters:
11109 @item psnr_log_version
11110 The version of the log file format. Will match @var{stats_version}.
11113 A comma separated list of the per-frame-pair parameters included in
11117 A description of each shown per-frame-pair parameter follows:
11121 sequential number of the input frame, starting from 1
11124 Mean Square Error pixel-by-pixel average difference of the compared
11125 frames, averaged over all the image components.
11127 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11128 Mean Square Error pixel-by-pixel average difference of the compared
11129 frames for the component specified by the suffix.
11131 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11132 Peak Signal to Noise ratio of the compared frames for the component
11133 specified by the suffix.
11135 @item max_avg, max_y, max_u, max_v
11136 Maximum allowed value for each channel, and average over all
11142 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11143 [main][ref] psnr="stats_file=stats.log" [out]
11146 On this example the input file being processed is compared with the
11147 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11148 is stored in @file{stats.log}.
11153 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11154 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11157 The pullup filter is designed to take advantage of future context in making
11158 its decisions. This filter is stateless in the sense that it does not lock
11159 onto a pattern to follow, but it instead looks forward to the following
11160 fields in order to identify matches and rebuild progressive frames.
11162 To produce content with an even framerate, insert the fps filter after
11163 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11164 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11166 The filter accepts the following options:
11173 These options set the amount of "junk" to ignore at the left, right, top, and
11174 bottom of the image, respectively. Left and right are in units of 8 pixels,
11175 while top and bottom are in units of 2 lines.
11176 The default is 8 pixels on each side.
11179 Set the strict breaks. Setting this option to 1 will reduce the chances of
11180 filter generating an occasional mismatched frame, but it may also cause an
11181 excessive number of frames to be dropped during high motion sequences.
11182 Conversely, setting it to -1 will make filter match fields more easily.
11183 This may help processing of video where there is slight blurring between
11184 the fields, but may also cause there to be interlaced frames in the output.
11185 Default value is @code{0}.
11188 Set the metric plane to use. It accepts the following values:
11194 Use chroma blue plane.
11197 Use chroma red plane.
11200 This option may be set to use chroma plane instead of the default luma plane
11201 for doing filter's computations. This may improve accuracy on very clean
11202 source material, but more likely will decrease accuracy, especially if there
11203 is chroma noise (rainbow effect) or any grayscale video.
11204 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11205 load and make pullup usable in realtime on slow machines.
11208 For best results (without duplicated frames in the output file) it is
11209 necessary to change the output frame rate. For example, to inverse
11210 telecine NTSC input:
11212 ffmpeg -i input -vf pullup -r 24000/1001 ...
11217 Change video quantization parameters (QP).
11219 The filter accepts the following option:
11223 Set expression for quantization parameter.
11226 The expression is evaluated through the eval API and can contain, among others,
11227 the following constants:
11231 1 if index is not 129, 0 otherwise.
11234 Sequentional index starting from -129 to 128.
11237 @subsection Examples
11241 Some equation like:
11249 Flush video frames from internal cache of frames into a random order.
11250 No frame is discarded.
11251 Inspired by @ref{frei0r} nervous filter.
11255 Set size in number of frames of internal cache, in range from @code{2} to
11256 @code{512}. Default is @code{30}.
11259 Set seed for random number generator, must be an integer included between
11260 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11261 less than @code{0}, the filter will try to use a good random seed on a
11265 @section readeia608
11267 Read closed captioning (EIA-608) information from the top lines of a video frame.
11269 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11270 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11271 with EIA-608 data (starting from 0). A description of each metadata value follows:
11274 @item lavfi.readeia608.X.cc
11275 The two bytes stored as EIA-608 data (printed in hexadecimal).
11277 @item lavfi.readeia608.X.line
11278 The number of the line on which the EIA-608 data was identified and read.
11281 This filter accepts the following options:
11285 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11288 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11291 Set minimal acceptable amplitude change for sync codes detection.
11292 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11295 Set the ratio of width reserved for sync code detection.
11296 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11299 Set the max peaks height difference for sync code detection.
11300 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11303 Set max peaks period difference for sync code detection.
11304 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11307 Set the first two max start code bits differences.
11308 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11311 Set the minimum ratio of bits height compared to 3rd start code bit.
11312 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11315 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11318 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11321 Enable checking the parity bit. In the event of a parity error, the filter will output
11322 @code{0x00} for that character. Default is false.
11325 @subsection Examples
11329 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11331 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
11337 Read vertical interval timecode (VITC) information from the top lines of a
11340 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11341 timecode value, if a valid timecode has been detected. Further metadata key
11342 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11343 timecode data has been found or not.
11345 This filter accepts the following options:
11349 Set the maximum number of lines to scan for VITC data. If the value is set to
11350 @code{-1} the full video frame is scanned. Default is @code{45}.
11353 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11354 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11357 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11358 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11361 @subsection Examples
11365 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11366 draw @code{--:--:--:--} as a placeholder:
11368 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11374 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11376 Destination pixel at position (X, Y) will be picked from source (x, y) position
11377 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11378 value for pixel will be used for destination pixel.
11380 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11381 will have Xmap/Ymap video stream dimensions.
11382 Xmap and Ymap input video streams are 16bit depth, single channel.
11384 @section removegrain
11386 The removegrain filter is a spatial denoiser for progressive video.
11390 Set mode for the first plane.
11393 Set mode for the second plane.
11396 Set mode for the third plane.
11399 Set mode for the fourth plane.
11402 Range of mode is from 0 to 24. Description of each mode follows:
11406 Leave input plane unchanged. Default.
11409 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11412 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11415 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11418 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11419 This is equivalent to a median filter.
11422 Line-sensitive clipping giving the minimal change.
11425 Line-sensitive clipping, intermediate.
11428 Line-sensitive clipping, intermediate.
11431 Line-sensitive clipping, intermediate.
11434 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11437 Replaces the target pixel with the closest neighbour.
11440 [1 2 1] horizontal and vertical kernel blur.
11446 Bob mode, interpolates top field from the line where the neighbours
11447 pixels are the closest.
11450 Bob mode, interpolates bottom field from the line where the neighbours
11451 pixels are the closest.
11454 Bob mode, interpolates top field. Same as 13 but with a more complicated
11455 interpolation formula.
11458 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11459 interpolation formula.
11462 Clips the pixel with the minimum and maximum of respectively the maximum and
11463 minimum of each pair of opposite neighbour pixels.
11466 Line-sensitive clipping using opposite neighbours whose greatest distance from
11467 the current pixel is minimal.
11470 Replaces the pixel with the average of its 8 neighbours.
11473 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11476 Clips pixels using the averages of opposite neighbour.
11479 Same as mode 21 but simpler and faster.
11482 Small edge and halo removal, but reputed useless.
11488 @section removelogo
11490 Suppress a TV station logo, using an image file to determine which
11491 pixels comprise the logo. It works by filling in the pixels that
11492 comprise the logo with neighboring pixels.
11494 The filter accepts the following options:
11498 Set the filter bitmap file, which can be any image format supported by
11499 libavformat. The width and height of the image file must match those of the
11500 video stream being processed.
11503 Pixels in the provided bitmap image with a value of zero are not
11504 considered part of the logo, non-zero pixels are considered part of
11505 the logo. If you use white (255) for the logo and black (0) for the
11506 rest, you will be safe. For making the filter bitmap, it is
11507 recommended to take a screen capture of a black frame with the logo
11508 visible, and then using a threshold filter followed by the erode
11509 filter once or twice.
11511 If needed, little splotches can be fixed manually. Remember that if
11512 logo pixels are not covered, the filter quality will be much
11513 reduced. Marking too many pixels as part of the logo does not hurt as
11514 much, but it will increase the amount of blurring needed to cover over
11515 the image and will destroy more information than necessary, and extra
11516 pixels will slow things down on a large logo.
11518 @section repeatfields
11520 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11521 fields based on its value.
11525 Reverse a video clip.
11527 Warning: This filter requires memory to buffer the entire clip, so trimming
11530 @subsection Examples
11534 Take the first 5 seconds of a clip, and reverse it.
11542 Rotate video by an arbitrary angle expressed in radians.
11544 The filter accepts the following options:
11546 A description of the optional parameters follows.
11549 Set an expression for the angle by which to rotate the input video
11550 clockwise, expressed as a number of radians. A negative value will
11551 result in a counter-clockwise rotation. By default it is set to "0".
11553 This expression is evaluated for each frame.
11556 Set the output width expression, default value is "iw".
11557 This expression is evaluated just once during configuration.
11560 Set the output height expression, default value is "ih".
11561 This expression is evaluated just once during configuration.
11564 Enable bilinear interpolation if set to 1, a value of 0 disables
11565 it. Default value is 1.
11568 Set the color used to fill the output area not covered by the rotated
11569 image. For the general syntax of this option, check the "Color" section in the
11570 ffmpeg-utils manual. If the special value "none" is selected then no
11571 background is printed (useful for example if the background is never shown).
11573 Default value is "black".
11576 The expressions for the angle and the output size can contain the
11577 following constants and functions:
11581 sequential number of the input frame, starting from 0. It is always NAN
11582 before the first frame is filtered.
11585 time in seconds of the input frame, it is set to 0 when the filter is
11586 configured. It is always NAN before the first frame is filtered.
11590 horizontal and vertical chroma subsample values. For example for the
11591 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11595 the input video width and height
11599 the output width and height, that is the size of the padded area as
11600 specified by the @var{width} and @var{height} expressions
11604 the minimal width/height required for completely containing the input
11605 video rotated by @var{a} radians.
11607 These are only available when computing the @option{out_w} and
11608 @option{out_h} expressions.
11611 @subsection Examples
11615 Rotate the input by PI/6 radians clockwise:
11621 Rotate the input by PI/6 radians counter-clockwise:
11627 Rotate the input by 45 degrees clockwise:
11633 Apply a constant rotation with period T, starting from an angle of PI/3:
11635 rotate=PI/3+2*PI*t/T
11639 Make the input video rotation oscillating with a period of T
11640 seconds and an amplitude of A radians:
11642 rotate=A*sin(2*PI/T*t)
11646 Rotate the video, output size is chosen so that the whole rotating
11647 input video is always completely contained in the output:
11649 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11653 Rotate the video, reduce the output size so that no background is ever
11656 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11660 @subsection Commands
11662 The filter supports the following commands:
11666 Set the angle expression.
11667 The command accepts the same syntax of the corresponding option.
11669 If the specified expression is not valid, it is kept at its current
11675 Apply Shape Adaptive Blur.
11677 The filter accepts the following options:
11680 @item luma_radius, lr
11681 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11682 value is 1.0. A greater value will result in a more blurred image, and
11683 in slower processing.
11685 @item luma_pre_filter_radius, lpfr
11686 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11689 @item luma_strength, ls
11690 Set luma maximum difference between pixels to still be considered, must
11691 be a value in the 0.1-100.0 range, default value is 1.0.
11693 @item chroma_radius, cr
11694 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11695 greater value will result in a more blurred image, and in slower
11698 @item chroma_pre_filter_radius, cpfr
11699 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11701 @item chroma_strength, cs
11702 Set chroma maximum difference between pixels to still be considered,
11703 must be a value in the -0.9-100.0 range.
11706 Each chroma option value, if not explicitly specified, is set to the
11707 corresponding luma option value.
11712 Scale (resize) the input video, using the libswscale library.
11714 The scale filter forces the output display aspect ratio to be the same
11715 of the input, by changing the output sample aspect ratio.
11717 If the input image format is different from the format requested by
11718 the next filter, the scale filter will convert the input to the
11721 @subsection Options
11722 The filter accepts the following options, or any of the options
11723 supported by the libswscale scaler.
11725 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11726 the complete list of scaler options.
11731 Set the output video dimension expression. Default value is the input
11734 If the value is 0, the input width is used for the output.
11736 If one of the values is -1, the scale filter will use a value that
11737 maintains the aspect ratio of the input image, calculated from the
11738 other specified dimension. If both of them are -1, the input size is
11741 If one of the values is -n with n > 1, the scale filter will also use a value
11742 that maintains the aspect ratio of the input image, calculated from the other
11743 specified dimension. After that it will, however, make sure that the calculated
11744 dimension is divisible by n and adjust the value if necessary.
11746 See below for the list of accepted constants for use in the dimension
11750 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11754 Only evaluate expressions once during the filter initialization or when a command is processed.
11757 Evaluate expressions for each incoming frame.
11761 Default value is @samp{init}.
11765 Set the interlacing mode. It accepts the following values:
11769 Force interlaced aware scaling.
11772 Do not apply interlaced scaling.
11775 Select interlaced aware scaling depending on whether the source frames
11776 are flagged as interlaced or not.
11779 Default value is @samp{0}.
11782 Set libswscale scaling flags. See
11783 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11784 complete list of values. If not explicitly specified the filter applies
11788 @item param0, param1
11789 Set libswscale input parameters for scaling algorithms that need them. See
11790 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11791 complete documentation. If not explicitly specified the filter applies
11797 Set the video size. For the syntax of this option, check the
11798 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11800 @item in_color_matrix
11801 @item out_color_matrix
11802 Set in/output YCbCr color space type.
11804 This allows the autodetected value to be overridden as well as allows forcing
11805 a specific value used for the output and encoder.
11807 If not specified, the color space type depends on the pixel format.
11813 Choose automatically.
11816 Format conforming to International Telecommunication Union (ITU)
11817 Recommendation BT.709.
11820 Set color space conforming to the United States Federal Communications
11821 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11824 Set color space conforming to:
11828 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11831 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11834 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11839 Set color space conforming to SMPTE ST 240:1999.
11844 Set in/output YCbCr sample range.
11846 This allows the autodetected value to be overridden as well as allows forcing
11847 a specific value used for the output and encoder. If not specified, the
11848 range depends on the pixel format. Possible values:
11852 Choose automatically.
11855 Set full range (0-255 in case of 8-bit luma).
11858 Set "MPEG" range (16-235 in case of 8-bit luma).
11861 @item force_original_aspect_ratio
11862 Enable decreasing or increasing output video width or height if necessary to
11863 keep the original aspect ratio. Possible values:
11867 Scale the video as specified and disable this feature.
11870 The output video dimensions will automatically be decreased if needed.
11873 The output video dimensions will automatically be increased if needed.
11877 One useful instance of this option is that when you know a specific device's
11878 maximum allowed resolution, you can use this to limit the output video to
11879 that, while retaining the aspect ratio. For example, device A allows
11880 1280x720 playback, and your video is 1920x800. Using this option (set it to
11881 decrease) and specifying 1280x720 to the command line makes the output
11884 Please note that this is a different thing than specifying -1 for @option{w}
11885 or @option{h}, you still need to specify the output resolution for this option
11890 The values of the @option{w} and @option{h} options are expressions
11891 containing the following constants:
11896 The input width and height
11900 These are the same as @var{in_w} and @var{in_h}.
11904 The output (scaled) width and height
11908 These are the same as @var{out_w} and @var{out_h}
11911 The same as @var{iw} / @var{ih}
11914 input sample aspect ratio
11917 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11921 horizontal and vertical input chroma subsample values. For example for the
11922 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11926 horizontal and vertical output chroma subsample values. For example for the
11927 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11930 @subsection Examples
11934 Scale the input video to a size of 200x100
11939 This is equivalent to:
11950 Specify a size abbreviation for the output size:
11955 which can also be written as:
11961 Scale the input to 2x:
11963 scale=w=2*iw:h=2*ih
11967 The above is the same as:
11969 scale=2*in_w:2*in_h
11973 Scale the input to 2x with forced interlaced scaling:
11975 scale=2*iw:2*ih:interl=1
11979 Scale the input to half size:
11981 scale=w=iw/2:h=ih/2
11985 Increase the width, and set the height to the same size:
11991 Seek Greek harmony:
11998 Increase the height, and set the width to 3/2 of the height:
12000 scale=w=3/2*oh:h=3/5*ih
12004 Increase the size, making the size a multiple of the chroma
12007 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12011 Increase the width to a maximum of 500 pixels,
12012 keeping the same aspect ratio as the input:
12014 scale=w='min(500\, iw*3/2):h=-1'
12018 @subsection Commands
12020 This filter supports the following commands:
12024 Set the output video dimension expression.
12025 The command accepts the same syntax of the corresponding option.
12027 If the specified expression is not valid, it is kept at its current
12033 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12034 format conversion on CUDA video frames. Setting the output width and height
12035 works in the same way as for the @var{scale} filter.
12037 The following additional options are accepted:
12040 The pixel format of the output CUDA frames. If set to the string "same" (the
12041 default), the input format will be kept. Note that automatic format negotiation
12042 and conversion is not yet supported for hardware frames
12045 The interpolation algorithm used for resizing. One of the following:
12052 @item cubic2p_bspline
12053 2-parameter cubic (B=1, C=0)
12055 @item cubic2p_catmullrom
12056 2-parameter cubic (B=0, C=1/2)
12058 @item cubic2p_b05c03
12059 2-parameter cubic (B=1/2, C=3/10)
12071 Scale (resize) the input video, based on a reference video.
12073 See the scale filter for available options, scale2ref supports the same but
12074 uses the reference video instead of the main input as basis.
12076 @subsection Examples
12080 Scale a subtitle stream to match the main video in size before overlaying
12082 'scale2ref[b][a];[a][b]overlay'
12086 @anchor{selectivecolor}
12087 @section selectivecolor
12089 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12090 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12091 by the "purity" of the color (that is, how saturated it already is).
12093 This filter is similar to the Adobe Photoshop Selective Color tool.
12095 The filter accepts the following options:
12098 @item correction_method
12099 Select color correction method.
12101 Available values are:
12104 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12107 Specified adjustments are relative to the original component value.
12109 Default is @code{absolute}.
12111 Adjustments for red pixels (pixels where the red component is the maximum)
12113 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12115 Adjustments for green pixels (pixels where the green component is the maximum)
12117 Adjustments for cyan pixels (pixels where the red component is the minimum)
12119 Adjustments for blue pixels (pixels where the blue component is the maximum)
12121 Adjustments for magenta pixels (pixels where the green component is the minimum)
12123 Adjustments for white pixels (pixels where all components are greater than 128)
12125 Adjustments for all pixels except pure black and pure white
12127 Adjustments for black pixels (pixels where all components are lesser than 128)
12129 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12132 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12133 4 space separated floating point adjustment values in the [-1,1] range,
12134 respectively to adjust the amount of cyan, magenta, yellow and black for the
12135 pixels of its range.
12137 @subsection Examples
12141 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12142 increase magenta by 27% in blue areas:
12144 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12148 Use a Photoshop selective color preset:
12150 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12154 @anchor{separatefields}
12155 @section separatefields
12157 The @code{separatefields} takes a frame-based video input and splits
12158 each frame into its components fields, producing a new half height clip
12159 with twice the frame rate and twice the frame count.
12161 This filter use field-dominance information in frame to decide which
12162 of each pair of fields to place first in the output.
12163 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12165 @section setdar, setsar
12167 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12170 This is done by changing the specified Sample (aka Pixel) Aspect
12171 Ratio, according to the following equation:
12173 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12176 Keep in mind that the @code{setdar} filter does not modify the pixel
12177 dimensions of the video frame. Also, the display aspect ratio set by
12178 this filter may be changed by later filters in the filterchain,
12179 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12182 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12183 the filter output video.
12185 Note that as a consequence of the application of this filter, the
12186 output display aspect ratio will change according to the equation
12189 Keep in mind that the sample aspect ratio set by the @code{setsar}
12190 filter may be changed by later filters in the filterchain, e.g. if
12191 another "setsar" or a "setdar" filter is applied.
12193 It accepts the following parameters:
12196 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12197 Set the aspect ratio used by the filter.
12199 The parameter can be a floating point number string, an expression, or
12200 a string of the form @var{num}:@var{den}, where @var{num} and
12201 @var{den} are the numerator and denominator of the aspect ratio. If
12202 the parameter is not specified, it is assumed the value "0".
12203 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12207 Set the maximum integer value to use for expressing numerator and
12208 denominator when reducing the expressed aspect ratio to a rational.
12209 Default value is @code{100}.
12213 The parameter @var{sar} is an expression containing
12214 the following constants:
12218 These are approximated values for the mathematical constants e
12219 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12222 The input width and height.
12225 These are the same as @var{w} / @var{h}.
12228 The input sample aspect ratio.
12231 The input display aspect ratio. It is the same as
12232 (@var{w} / @var{h}) * @var{sar}.
12235 Horizontal and vertical chroma subsample values. For example, for the
12236 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12239 @subsection Examples
12244 To change the display aspect ratio to 16:9, specify one of the following:
12251 To change the sample aspect ratio to 10:11, specify:
12257 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12258 1000 in the aspect ratio reduction, use the command:
12260 setdar=ratio=16/9:max=1000
12268 Force field for the output video frame.
12270 The @code{setfield} filter marks the interlace type field for the
12271 output frames. It does not change the input frame, but only sets the
12272 corresponding property, which affects how the frame is treated by
12273 following filters (e.g. @code{fieldorder} or @code{yadif}).
12275 The filter accepts the following options:
12280 Available values are:
12284 Keep the same field property.
12287 Mark the frame as bottom-field-first.
12290 Mark the frame as top-field-first.
12293 Mark the frame as progressive.
12299 Show a line containing various information for each input video frame.
12300 The input video is not modified.
12302 The shown line contains a sequence of key/value pairs of the form
12303 @var{key}:@var{value}.
12305 The following values are shown in the output:
12309 The (sequential) number of the input frame, starting from 0.
12312 The Presentation TimeStamp of the input frame, expressed as a number of
12313 time base units. The time base unit depends on the filter input pad.
12316 The Presentation TimeStamp of the input frame, expressed as a number of
12320 The position of the frame in the input stream, or -1 if this information is
12321 unavailable and/or meaningless (for example in case of synthetic video).
12324 The pixel format name.
12327 The sample aspect ratio of the input frame, expressed in the form
12328 @var{num}/@var{den}.
12331 The size of the input frame. For the syntax of this option, check the
12332 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12335 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12336 for bottom field first).
12339 This is 1 if the frame is a key frame, 0 otherwise.
12342 The picture type of the input frame ("I" for an I-frame, "P" for a
12343 P-frame, "B" for a B-frame, or "?" for an unknown type).
12344 Also refer to the documentation of the @code{AVPictureType} enum and of
12345 the @code{av_get_picture_type_char} function defined in
12346 @file{libavutil/avutil.h}.
12349 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12351 @item plane_checksum
12352 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12353 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12356 @section showpalette
12358 Displays the 256 colors palette of each frame. This filter is only relevant for
12359 @var{pal8} pixel format frames.
12361 It accepts the following option:
12365 Set the size of the box used to represent one palette color entry. Default is
12366 @code{30} (for a @code{30x30} pixel box).
12369 @section shuffleframes
12371 Reorder and/or duplicate and/or drop video frames.
12373 It accepts the following parameters:
12377 Set the destination indexes of input frames.
12378 This is space or '|' separated list of indexes that maps input frames to output
12379 frames. Number of indexes also sets maximal value that each index may have.
12380 '-1' index have special meaning and that is to drop frame.
12383 The first frame has the index 0. The default is to keep the input unchanged.
12385 @subsection Examples
12389 Swap second and third frame of every three frames of the input:
12391 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12395 Swap 10th and 1st frame of every ten frames of the input:
12397 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12401 @section shuffleplanes
12403 Reorder and/or duplicate video planes.
12405 It accepts the following parameters:
12410 The index of the input plane to be used as the first output plane.
12413 The index of the input plane to be used as the second output plane.
12416 The index of the input plane to be used as the third output plane.
12419 The index of the input plane to be used as the fourth output plane.
12423 The first plane has the index 0. The default is to keep the input unchanged.
12425 @subsection Examples
12429 Swap the second and third planes of the input:
12431 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12435 @anchor{signalstats}
12436 @section signalstats
12437 Evaluate various visual metrics that assist in determining issues associated
12438 with the digitization of analog video media.
12440 By default the filter will log these metadata values:
12444 Display the minimal Y value contained within the input frame. Expressed in
12448 Display the Y value at the 10% percentile within the input frame. Expressed in
12452 Display the average Y value within the input frame. Expressed in range of
12456 Display the Y value at the 90% percentile within the input frame. Expressed in
12460 Display the maximum Y value contained within the input frame. Expressed in
12464 Display the minimal U value contained within the input frame. Expressed in
12468 Display the U value at the 10% percentile within the input frame. Expressed in
12472 Display the average U value within the input frame. Expressed in range of
12476 Display the U value at the 90% percentile within the input frame. Expressed in
12480 Display the maximum U value contained within the input frame. Expressed in
12484 Display the minimal V value contained within the input frame. Expressed in
12488 Display the V value at the 10% percentile within the input frame. Expressed in
12492 Display the average V value within the input frame. Expressed in range of
12496 Display the V value at the 90% percentile within the input frame. Expressed in
12500 Display the maximum V value contained within the input frame. Expressed in
12504 Display the minimal saturation value contained within the input frame.
12505 Expressed in range of [0-~181.02].
12508 Display the saturation value at the 10% percentile within the input frame.
12509 Expressed in range of [0-~181.02].
12512 Display the average saturation value within the input frame. Expressed in range
12516 Display the saturation value at the 90% percentile within the input frame.
12517 Expressed in range of [0-~181.02].
12520 Display the maximum saturation value contained within the input frame.
12521 Expressed in range of [0-~181.02].
12524 Display the median value for hue within the input frame. Expressed in range of
12528 Display the average value for hue within the input frame. Expressed in range of
12532 Display the average of sample value difference between all values of the Y
12533 plane in the current frame and corresponding values of the previous input frame.
12534 Expressed in range of [0-255].
12537 Display the average of sample value difference between all values of the U
12538 plane in the current frame and corresponding values of the previous input frame.
12539 Expressed in range of [0-255].
12542 Display the average of sample value difference between all values of the V
12543 plane in the current frame and corresponding values of the previous input frame.
12544 Expressed in range of [0-255].
12547 Display bit depth of Y plane in current frame.
12548 Expressed in range of [0-16].
12551 Display bit depth of U plane in current frame.
12552 Expressed in range of [0-16].
12555 Display bit depth of V plane in current frame.
12556 Expressed in range of [0-16].
12559 The filter accepts the following options:
12565 @option{stat} specify an additional form of image analysis.
12566 @option{out} output video with the specified type of pixel highlighted.
12568 Both options accept the following values:
12572 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12573 unlike the neighboring pixels of the same field. Examples of temporal outliers
12574 include the results of video dropouts, head clogs, or tape tracking issues.
12577 Identify @var{vertical line repetition}. Vertical line repetition includes
12578 similar rows of pixels within a frame. In born-digital video vertical line
12579 repetition is common, but this pattern is uncommon in video digitized from an
12580 analog source. When it occurs in video that results from the digitization of an
12581 analog source it can indicate concealment from a dropout compensator.
12584 Identify pixels that fall outside of legal broadcast range.
12588 Set the highlight color for the @option{out} option. The default color is
12592 @subsection Examples
12596 Output data of various video metrics:
12598 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12602 Output specific data about the minimum and maximum values of the Y plane per frame:
12604 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12608 Playback video while highlighting pixels that are outside of broadcast range in red.
12610 ffplay example.mov -vf signalstats="out=brng:color=red"
12614 Playback video with signalstats metadata drawn over the frame.
12616 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12619 The contents of signalstat_drawtext.txt used in the command are:
12622 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12623 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12624 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12625 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12633 Calculates the MPEG-7 Video Signature. The filter can handle more than one
12634 input. In this case the matching between the inputs can be calculated additionally.
12635 The filter always passes through the first input. The signature of each stream can
12636 be written into a file.
12638 It accepts the following options:
12642 Enable or disable the matching process.
12644 Available values are:
12648 Disable the calculation of a matching (default).
12650 Calculate the matching for the whole video and output whether the whole video
12651 matches or only parts.
12653 Calculate only until a matching is found or the video ends. Should be faster in
12658 Set the number of inputs. The option value must be a non negative integer.
12659 Default value is 1.
12662 Set the path to which the output is written. If there is more than one input,
12663 the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12664 integer), that will be replaced with the input number. If no filename is
12665 specified, no output will be written. This is the default.
12668 Choose the output format.
12670 Available values are:
12674 Use the specified binary representation (default).
12676 Use the specified xml representation.
12680 Set threshold to detect one word as similar. The option value must be an integer
12681 greater than zero. The default value is 9000.
12684 Set threshold to detect all words as similar. The option value must be an integer
12685 greater than zero. The default value is 60000.
12688 Set threshold to detect frames as similar. The option value must be an integer
12689 greater than zero. The default value is 116.
12692 Set the minimum length of a sequence in frames to recognize it as matching
12693 sequence. The option value must be a non negative integer value.
12694 The default value is 0.
12697 Set the minimum relation, that matching frames to all frames must have.
12698 The option value must be a double value between 0 and 1. The default value is 0.5.
12701 @subsection Examples
12705 To calculate the signature of an input video and store it in signature.bin:
12707 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12711 To detect whether two videos match and store the signatures in XML format in
12712 signature0.xml and signature1.xml:
12714 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 -
12722 Blur the input video without impacting the outlines.
12724 It accepts the following options:
12727 @item luma_radius, lr
12728 Set the luma radius. The option value must be a float number in
12729 the range [0.1,5.0] that specifies the variance of the gaussian filter
12730 used to blur the image (slower if larger). Default value is 1.0.
12732 @item luma_strength, ls
12733 Set the luma strength. The option value must be a float number
12734 in the range [-1.0,1.0] that configures the blurring. A value included
12735 in [0.0,1.0] will blur the image whereas a value included in
12736 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12738 @item luma_threshold, lt
12739 Set the luma threshold used as a coefficient to determine
12740 whether a pixel should be blurred or not. The option value must be an
12741 integer in the range [-30,30]. A value of 0 will filter all the image,
12742 a value included in [0,30] will filter flat areas and a value included
12743 in [-30,0] will filter edges. Default value is 0.
12745 @item chroma_radius, cr
12746 Set the chroma radius. The option value must be a float number in
12747 the range [0.1,5.0] that specifies the variance of the gaussian filter
12748 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12750 @item chroma_strength, cs
12751 Set the chroma strength. The option value must be a float number
12752 in the range [-1.0,1.0] that configures the blurring. A value included
12753 in [0.0,1.0] will blur the image whereas a value included in
12754 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12756 @item chroma_threshold, ct
12757 Set the chroma threshold used as a coefficient to determine
12758 whether a pixel should be blurred or not. The option value must be an
12759 integer in the range [-30,30]. A value of 0 will filter all the image,
12760 a value included in [0,30] will filter flat areas and a value included
12761 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12764 If a chroma option is not explicitly set, the corresponding luma value
12769 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12771 This filter takes in input two input videos, the first input is
12772 considered the "main" source and is passed unchanged to the
12773 output. The second input is used as a "reference" video for computing
12776 Both video inputs must have the same resolution and pixel format for
12777 this filter to work correctly. Also it assumes that both inputs
12778 have the same number of frames, which are compared one by one.
12780 The filter stores the calculated SSIM of each frame.
12782 The description of the accepted parameters follows.
12785 @item stats_file, f
12786 If specified the filter will use the named file to save the SSIM of
12787 each individual frame. When filename equals "-" the data is sent to
12791 The file printed if @var{stats_file} is selected, contains a sequence of
12792 key/value pairs of the form @var{key}:@var{value} for each compared
12795 A description of each shown parameter follows:
12799 sequential number of the input frame, starting from 1
12801 @item Y, U, V, R, G, B
12802 SSIM of the compared frames for the component specified by the suffix.
12805 SSIM of the compared frames for the whole frame.
12808 Same as above but in dB representation.
12813 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12814 [main][ref] ssim="stats_file=stats.log" [out]
12817 On this example the input file being processed is compared with the
12818 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12819 is stored in @file{stats.log}.
12821 Another example with both psnr and ssim at same time:
12823 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12828 Convert between different stereoscopic image formats.
12830 The filters accept the following options:
12834 Set stereoscopic image format of input.
12836 Available values for input image formats are:
12839 side by side parallel (left eye left, right eye right)
12842 side by side crosseye (right eye left, left eye right)
12845 side by side parallel with half width resolution
12846 (left eye left, right eye right)
12849 side by side crosseye with half width resolution
12850 (right eye left, left eye right)
12853 above-below (left eye above, right eye below)
12856 above-below (right eye above, left eye below)
12859 above-below with half height resolution
12860 (left eye above, right eye below)
12863 above-below with half height resolution
12864 (right eye above, left eye below)
12867 alternating frames (left eye first, right eye second)
12870 alternating frames (right eye first, left eye second)
12873 interleaved rows (left eye has top row, right eye starts on next row)
12876 interleaved rows (right eye has top row, left eye starts on next row)
12879 interleaved columns, left eye first
12882 interleaved columns, right eye first
12884 Default value is @samp{sbsl}.
12888 Set stereoscopic image format of output.
12892 side by side parallel (left eye left, right eye right)
12895 side by side crosseye (right eye left, left eye right)
12898 side by side parallel with half width resolution
12899 (left eye left, right eye right)
12902 side by side crosseye with half width resolution
12903 (right eye left, left eye right)
12906 above-below (left eye above, right eye below)
12909 above-below (right eye above, left eye below)
12912 above-below with half height resolution
12913 (left eye above, right eye below)
12916 above-below with half height resolution
12917 (right eye above, left eye below)
12920 alternating frames (left eye first, right eye second)
12923 alternating frames (right eye first, left eye second)
12926 interleaved rows (left eye has top row, right eye starts on next row)
12929 interleaved rows (right eye has top row, left eye starts on next row)
12932 anaglyph red/blue gray
12933 (red filter on left eye, blue filter on right eye)
12936 anaglyph red/green gray
12937 (red filter on left eye, green filter on right eye)
12940 anaglyph red/cyan gray
12941 (red filter on left eye, cyan filter on right eye)
12944 anaglyph red/cyan half colored
12945 (red filter on left eye, cyan filter on right eye)
12948 anaglyph red/cyan color
12949 (red filter on left eye, cyan filter on right eye)
12952 anaglyph red/cyan color optimized with the least squares projection of dubois
12953 (red filter on left eye, cyan filter on right eye)
12956 anaglyph green/magenta gray
12957 (green filter on left eye, magenta filter on right eye)
12960 anaglyph green/magenta half colored
12961 (green filter on left eye, magenta filter on right eye)
12964 anaglyph green/magenta colored
12965 (green filter on left eye, magenta filter on right eye)
12968 anaglyph green/magenta color optimized with the least squares projection of dubois
12969 (green filter on left eye, magenta filter on right eye)
12972 anaglyph yellow/blue gray
12973 (yellow filter on left eye, blue filter on right eye)
12976 anaglyph yellow/blue half colored
12977 (yellow filter on left eye, blue filter on right eye)
12980 anaglyph yellow/blue colored
12981 (yellow filter on left eye, blue filter on right eye)
12984 anaglyph yellow/blue color optimized with the least squares projection of dubois
12985 (yellow filter on left eye, blue filter on right eye)
12988 mono output (left eye only)
12991 mono output (right eye only)
12994 checkerboard, left eye first
12997 checkerboard, right eye first
13000 interleaved columns, left eye first
13003 interleaved columns, right eye first
13009 Default value is @samp{arcd}.
13012 @subsection Examples
13016 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13022 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13028 @section streamselect, astreamselect
13029 Select video or audio streams.
13031 The filter accepts the following options:
13035 Set number of inputs. Default is 2.
13038 Set input indexes to remap to outputs.
13041 @subsection Commands
13043 The @code{streamselect} and @code{astreamselect} filter supports the following
13048 Set input indexes to remap to outputs.
13051 @subsection Examples
13055 Select first 5 seconds 1st stream and rest of time 2nd stream:
13057 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13061 Same as above, but for audio:
13063 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13068 Apply sobel operator to input video stream.
13070 The filter accepts the following option:
13074 Set which planes will be processed, unprocessed planes will be copied.
13075 By default value 0xf, all planes will be processed.
13078 Set value which will be multiplied with filtered result.
13081 Set value which will be added to filtered result.
13087 Apply a simple postprocessing filter that compresses and decompresses the image
13088 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13089 and average the results.
13091 The filter accepts the following options:
13095 Set quality. This option defines the number of levels for averaging. It accepts
13096 an integer in the range 0-6. If set to @code{0}, the filter will have no
13097 effect. A value of @code{6} means the higher quality. For each increment of
13098 that value the speed drops by a factor of approximately 2. Default value is
13102 Force a constant quantization parameter. If not set, the filter will use the QP
13103 from the video stream (if available).
13106 Set thresholding mode. Available modes are:
13110 Set hard thresholding (default).
13112 Set soft thresholding (better de-ringing effect, but likely blurrier).
13115 @item use_bframe_qp
13116 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13117 option may cause flicker since the B-Frames have often larger QP. Default is
13118 @code{0} (not enabled).
13124 Draw subtitles on top of input video using the libass library.
13126 To enable compilation of this filter you need to configure FFmpeg with
13127 @code{--enable-libass}. This filter also requires a build with libavcodec and
13128 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13129 Alpha) subtitles format.
13131 The filter accepts the following options:
13135 Set the filename of the subtitle file to read. It must be specified.
13137 @item original_size
13138 Specify the size of the original video, the video for which the ASS file
13139 was composed. For the syntax of this option, check the
13140 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13141 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13142 correctly scale the fonts if the aspect ratio has been changed.
13145 Set a directory path containing fonts that can be used by the filter.
13146 These fonts will be used in addition to whatever the font provider uses.
13149 Set subtitles input character encoding. @code{subtitles} filter only. Only
13150 useful if not UTF-8.
13152 @item stream_index, si
13153 Set subtitles stream index. @code{subtitles} filter only.
13156 Override default style or script info parameters of the subtitles. It accepts a
13157 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13160 If the first key is not specified, it is assumed that the first value
13161 specifies the @option{filename}.
13163 For example, to render the file @file{sub.srt} on top of the input
13164 video, use the command:
13169 which is equivalent to:
13171 subtitles=filename=sub.srt
13174 To render the default subtitles stream from file @file{video.mkv}, use:
13176 subtitles=video.mkv
13179 To render the second subtitles stream from that file, use:
13181 subtitles=video.mkv:si=1
13184 To make the subtitles stream from @file{sub.srt} appear in transparent green
13185 @code{DejaVu Serif}, use:
13187 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13190 @section super2xsai
13192 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13193 Interpolate) pixel art scaling algorithm.
13195 Useful for enlarging pixel art images without reducing sharpness.
13199 Swap two rectangular objects in video.
13201 This filter accepts the following options:
13211 Set 1st rect x coordinate.
13214 Set 1st rect y coordinate.
13217 Set 2nd rect x coordinate.
13220 Set 2nd rect y coordinate.
13222 All expressions are evaluated once for each frame.
13225 The all options are expressions containing the following constants:
13230 The input width and height.
13233 same as @var{w} / @var{h}
13236 input sample aspect ratio
13239 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13242 The number of the input frame, starting from 0.
13245 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13248 the position in the file of the input frame, NAN if unknown
13256 Apply telecine process to the video.
13258 This filter accepts the following options:
13267 The default value is @code{top}.
13271 A string of numbers representing the pulldown pattern you wish to apply.
13272 The default value is @code{23}.
13276 Some typical patterns:
13281 24p: 2332 (preferred)
13288 24p: 222222222223 ("Euro pulldown")
13295 Apply threshold effect to video stream.
13297 This filter needs four video streams to perform thresholding.
13298 First stream is stream we are filtering.
13299 Second stream is holding threshold values, third stream is holding min values,
13300 and last, fourth stream is holding max values.
13302 The filter accepts the following option:
13306 Set which planes will be processed, unprocessed planes will be copied.
13307 By default value 0xf, all planes will be processed.
13310 For example if first stream pixel's component value is less then threshold value
13311 of pixel component from 2nd threshold stream, third stream value will picked,
13312 otherwise fourth stream pixel component value will be picked.
13314 Using color source filter one can perform various types of thresholding:
13316 @subsection Examples
13320 Binary threshold, using gray color as threshold:
13322 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13326 Inverted binary threshold, using gray color as threshold:
13328 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13332 Truncate binary threshold, using gray color as threshold:
13334 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13338 Threshold to zero, using gray color as threshold:
13340 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13344 Inverted threshold to zero, using gray color as threshold:
13346 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13351 Select the most representative frame in a given sequence of consecutive frames.
13353 The filter accepts the following options:
13357 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13358 will pick one of them, and then handle the next batch of @var{n} frames until
13359 the end. Default is @code{100}.
13362 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13363 value will result in a higher memory usage, so a high value is not recommended.
13365 @subsection Examples
13369 Extract one picture each 50 frames:
13375 Complete example of a thumbnail creation with @command{ffmpeg}:
13377 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13383 Tile several successive frames together.
13385 The filter accepts the following options:
13390 Set the grid size (i.e. the number of lines and columns). For the syntax of
13391 this option, check the
13392 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13395 Set the maximum number of frames to render in the given area. It must be less
13396 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13397 the area will be used.
13400 Set the outer border margin in pixels.
13403 Set the inner border thickness (i.e. the number of pixels between frames). For
13404 more advanced padding options (such as having different values for the edges),
13405 refer to the pad video filter.
13408 Specify the color of the unused area. For the syntax of this option, check the
13409 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13413 @subsection Examples
13417 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13419 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13421 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13422 duplicating each output frame to accommodate the originally detected frame
13426 Display @code{5} pictures in an area of @code{3x2} frames,
13427 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13428 mixed flat and named options:
13430 tile=3x2:nb_frames=5:padding=7:margin=2
13434 @section tinterlace
13436 Perform various types of temporal field interlacing.
13438 Frames are counted starting from 1, so the first input frame is
13441 The filter accepts the following options:
13446 Specify the mode of the interlacing. This option can also be specified
13447 as a value alone. See below for a list of values for this option.
13449 Available values are:
13453 Move odd frames into the upper field, even into the lower field,
13454 generating a double height frame at half frame rate.
13458 Frame 1 Frame 2 Frame 3 Frame 4
13460 11111 22222 33333 44444
13461 11111 22222 33333 44444
13462 11111 22222 33333 44444
13463 11111 22222 33333 44444
13477 Only output odd frames, even frames are dropped, generating a frame with
13478 unchanged height at half frame rate.
13483 Frame 1 Frame 2 Frame 3 Frame 4
13485 11111 22222 33333 44444
13486 11111 22222 33333 44444
13487 11111 22222 33333 44444
13488 11111 22222 33333 44444
13498 Only output even frames, odd frames are dropped, generating a frame with
13499 unchanged height at half frame rate.
13504 Frame 1 Frame 2 Frame 3 Frame 4
13506 11111 22222 33333 44444
13507 11111 22222 33333 44444
13508 11111 22222 33333 44444
13509 11111 22222 33333 44444
13519 Expand each frame to full height, but pad alternate lines with black,
13520 generating a frame with double height at the same input frame rate.
13525 Frame 1 Frame 2 Frame 3 Frame 4
13527 11111 22222 33333 44444
13528 11111 22222 33333 44444
13529 11111 22222 33333 44444
13530 11111 22222 33333 44444
13533 11111 ..... 33333 .....
13534 ..... 22222 ..... 44444
13535 11111 ..... 33333 .....
13536 ..... 22222 ..... 44444
13537 11111 ..... 33333 .....
13538 ..... 22222 ..... 44444
13539 11111 ..... 33333 .....
13540 ..... 22222 ..... 44444
13544 @item interleave_top, 4
13545 Interleave the upper field from odd frames with the lower field from
13546 even frames, generating a frame with unchanged height at half frame rate.
13551 Frame 1 Frame 2 Frame 3 Frame 4
13553 11111<- 22222 33333<- 44444
13554 11111 22222<- 33333 44444<-
13555 11111<- 22222 33333<- 44444
13556 11111 22222<- 33333 44444<-
13566 @item interleave_bottom, 5
13567 Interleave the lower field from odd frames with the upper field from
13568 even frames, generating a frame with unchanged height at half frame rate.
13573 Frame 1 Frame 2 Frame 3 Frame 4
13575 11111 22222<- 33333 44444<-
13576 11111<- 22222 33333<- 44444
13577 11111 22222<- 33333 44444<-
13578 11111<- 22222 33333<- 44444
13588 @item interlacex2, 6
13589 Double frame rate with unchanged height. Frames are inserted each
13590 containing the second temporal field from the previous input frame and
13591 the first temporal field from the next input frame. This mode relies on
13592 the top_field_first flag. Useful for interlaced video displays with no
13593 field synchronisation.
13598 Frame 1 Frame 2 Frame 3 Frame 4
13600 11111 22222 33333 44444
13601 11111 22222 33333 44444
13602 11111 22222 33333 44444
13603 11111 22222 33333 44444
13606 11111 22222 22222 33333 33333 44444 44444
13607 11111 11111 22222 22222 33333 33333 44444
13608 11111 22222 22222 33333 33333 44444 44444
13609 11111 11111 22222 22222 33333 33333 44444
13614 Move odd frames into the upper field, even into the lower field,
13615 generating a double height frame at same frame rate.
13620 Frame 1 Frame 2 Frame 3 Frame 4
13622 11111 22222 33333 44444
13623 11111 22222 33333 44444
13624 11111 22222 33333 44444
13625 11111 22222 33333 44444
13628 11111 33333 33333 55555
13629 22222 22222 44444 44444
13630 11111 33333 33333 55555
13631 22222 22222 44444 44444
13632 11111 33333 33333 55555
13633 22222 22222 44444 44444
13634 11111 33333 33333 55555
13635 22222 22222 44444 44444
13640 Numeric values are deprecated but are accepted for backward
13641 compatibility reasons.
13643 Default mode is @code{merge}.
13646 Specify flags influencing the filter process.
13648 Available value for @var{flags} is:
13651 @item low_pass_filter, vlfp
13652 Enable vertical low-pass filtering in the filter.
13653 Vertical low-pass filtering is required when creating an interlaced
13654 destination from a progressive source which contains high-frequency
13655 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13658 Vertical low-pass filtering can only be enabled for @option{mode}
13659 @var{interleave_top} and @var{interleave_bottom}.
13666 Transpose rows with columns in the input video and optionally flip it.
13668 It accepts the following parameters:
13673 Specify the transposition direction.
13675 Can assume the following values:
13677 @item 0, 4, cclock_flip
13678 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13686 Rotate by 90 degrees clockwise, that is:
13694 Rotate by 90 degrees counterclockwise, that is:
13701 @item 3, 7, clock_flip
13702 Rotate by 90 degrees clockwise and vertically flip, that is:
13710 For values between 4-7, the transposition is only done if the input
13711 video geometry is portrait and not landscape. These values are
13712 deprecated, the @code{passthrough} option should be used instead.
13714 Numerical values are deprecated, and should be dropped in favor of
13715 symbolic constants.
13718 Do not apply the transposition if the input geometry matches the one
13719 specified by the specified value. It accepts the following values:
13722 Always apply transposition.
13724 Preserve portrait geometry (when @var{height} >= @var{width}).
13726 Preserve landscape geometry (when @var{width} >= @var{height}).
13729 Default value is @code{none}.
13732 For example to rotate by 90 degrees clockwise and preserve portrait
13735 transpose=dir=1:passthrough=portrait
13738 The command above can also be specified as:
13740 transpose=1:portrait
13744 Trim the input so that the output contains one continuous subpart of the input.
13746 It accepts the following parameters:
13749 Specify the time of the start of the kept section, i.e. the frame with the
13750 timestamp @var{start} will be the first frame in the output.
13753 Specify the time of the first frame that will be dropped, i.e. the frame
13754 immediately preceding the one with the timestamp @var{end} will be the last
13755 frame in the output.
13758 This is the same as @var{start}, except this option sets the start timestamp
13759 in timebase units instead of seconds.
13762 This is the same as @var{end}, except this option sets the end timestamp
13763 in timebase units instead of seconds.
13766 The maximum duration of the output in seconds.
13769 The number of the first frame that should be passed to the output.
13772 The number of the first frame that should be dropped.
13775 @option{start}, @option{end}, and @option{duration} are expressed as time
13776 duration specifications; see
13777 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13778 for the accepted syntax.
13780 Note that the first two sets of the start/end options and the @option{duration}
13781 option look at the frame timestamp, while the _frame variants simply count the
13782 frames that pass through the filter. Also note that this filter does not modify
13783 the timestamps. If you wish for the output timestamps to start at zero, insert a
13784 setpts filter after the trim filter.
13786 If multiple start or end options are set, this filter tries to be greedy and
13787 keep all the frames that match at least one of the specified constraints. To keep
13788 only the part that matches all the constraints at once, chain multiple trim
13791 The defaults are such that all the input is kept. So it is possible to set e.g.
13792 just the end values to keep everything before the specified time.
13797 Drop everything except the second minute of input:
13799 ffmpeg -i INPUT -vf trim=60:120
13803 Keep only the first second:
13805 ffmpeg -i INPUT -vf trim=duration=1
13814 Sharpen or blur the input video.
13816 It accepts the following parameters:
13819 @item luma_msize_x, lx
13820 Set the luma matrix horizontal size. It must be an odd integer between
13821 3 and 23. The default value is 5.
13823 @item luma_msize_y, ly
13824 Set the luma matrix vertical size. It must be an odd integer between 3
13825 and 23. The default value is 5.
13827 @item luma_amount, la
13828 Set the luma effect strength. It must be a floating point number, reasonable
13829 values lay between -1.5 and 1.5.
13831 Negative values will blur the input video, while positive values will
13832 sharpen it, a value of zero will disable the effect.
13834 Default value is 1.0.
13836 @item chroma_msize_x, cx
13837 Set the chroma matrix horizontal size. It must be an odd integer
13838 between 3 and 23. The default value is 5.
13840 @item chroma_msize_y, cy
13841 Set the chroma matrix vertical size. It must be an odd integer
13842 between 3 and 23. The default value is 5.
13844 @item chroma_amount, ca
13845 Set the chroma effect strength. It must be a floating point number, reasonable
13846 values lay between -1.5 and 1.5.
13848 Negative values will blur the input video, while positive values will
13849 sharpen it, a value of zero will disable the effect.
13851 Default value is 0.0.
13854 If set to 1, specify using OpenCL capabilities, only available if
13855 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13859 All parameters are optional and default to the equivalent of the
13860 string '5:5:1.0:5:5:0.0'.
13862 @subsection Examples
13866 Apply strong luma sharpen effect:
13868 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13872 Apply a strong blur of both luma and chroma parameters:
13874 unsharp=7:7:-2:7:7:-2
13880 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13881 the image at several (or - in the case of @option{quality} level @code{8} - all)
13882 shifts and average the results.
13884 The way this differs from the behavior of spp is that uspp actually encodes &
13885 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13886 DCT similar to MJPEG.
13888 The filter accepts the following options:
13892 Set quality. This option defines the number of levels for averaging. It accepts
13893 an integer in the range 0-8. If set to @code{0}, the filter will have no
13894 effect. A value of @code{8} means the higher quality. For each increment of
13895 that value the speed drops by a factor of approximately 2. Default value is
13899 Force a constant quantization parameter. If not set, the filter will use the QP
13900 from the video stream (if available).
13903 @section vaguedenoiser
13905 Apply a wavelet based denoiser.
13907 It transforms each frame from the video input into the wavelet domain,
13908 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
13909 the obtained coefficients. It does an inverse wavelet transform after.
13910 Due to wavelet properties, it should give a nice smoothed result, and
13911 reduced noise, without blurring picture features.
13913 This filter accepts the following options:
13917 The filtering strength. The higher, the more filtered the video will be.
13918 Hard thresholding can use a higher threshold than soft thresholding
13919 before the video looks overfiltered.
13922 The filtering method the filter will use.
13924 It accepts the following values:
13927 All values under the threshold will be zeroed.
13930 All values under the threshold will be zeroed. All values above will be
13931 reduced by the threshold.
13934 Scales or nullifies coefficients - intermediary between (more) soft and
13935 (less) hard thresholding.
13939 Number of times, the wavelet will decompose the picture. Picture can't
13940 be decomposed beyond a particular point (typically, 8 for a 640x480
13941 frame - as 2^9 = 512 > 480)
13944 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
13947 A list of the planes to process. By default all planes are processed.
13950 @section vectorscope
13952 Display 2 color component values in the two dimensional graph (which is called
13955 This filter accepts the following options:
13959 Set vectorscope mode.
13961 It accepts the following values:
13964 Gray values are displayed on graph, higher brightness means more pixels have
13965 same component color value on location in graph. This is the default mode.
13968 Gray values are displayed on graph. Surrounding pixels values which are not
13969 present in video frame are drawn in gradient of 2 color components which are
13970 set by option @code{x} and @code{y}. The 3rd color component is static.
13973 Actual color components values present in video frame are displayed on graph.
13976 Similar as color2 but higher frequency of same values @code{x} and @code{y}
13977 on graph increases value of another color component, which is luminance by
13978 default values of @code{x} and @code{y}.
13981 Actual colors present in video frame are displayed on graph. If two different
13982 colors map to same position on graph then color with higher value of component
13983 not present in graph is picked.
13986 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
13987 component picked from radial gradient.
13991 Set which color component will be represented on X-axis. Default is @code{1}.
13994 Set which color component will be represented on Y-axis. Default is @code{2}.
13997 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
13998 of color component which represents frequency of (X, Y) location in graph.
14003 No envelope, this is default.
14006 Instant envelope, even darkest single pixel will be clearly highlighted.
14009 Hold maximum and minimum values presented in graph over time. This way you
14010 can still spot out of range values without constantly looking at vectorscope.
14013 Peak and instant envelope combined together.
14017 Set what kind of graticule to draw.
14025 Set graticule opacity.
14028 Set graticule flags.
14032 Draw graticule for white point.
14035 Draw graticule for black point.
14038 Draw color points short names.
14042 Set background opacity.
14044 @item lthreshold, l
14045 Set low threshold for color component not represented on X or Y axis.
14046 Values lower than this value will be ignored. Default is 0.
14047 Note this value is multiplied with actual max possible value one pixel component
14048 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14051 @item hthreshold, h
14052 Set high threshold for color component not represented on X or Y axis.
14053 Values higher than this value will be ignored. Default is 1.
14054 Note this value is multiplied with actual max possible value one pixel component
14055 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14056 is 0.9 * 255 = 230.
14058 @item colorspace, c
14059 Set what kind of colorspace to use when drawing graticule.
14068 @anchor{vidstabdetect}
14069 @section vidstabdetect
14071 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14072 @ref{vidstabtransform} for pass 2.
14074 This filter generates a file with relative translation and rotation
14075 transform information about subsequent frames, which is then used by
14076 the @ref{vidstabtransform} filter.
14078 To enable compilation of this filter you need to configure FFmpeg with
14079 @code{--enable-libvidstab}.
14081 This filter accepts the following options:
14085 Set the path to the file used to write the transforms information.
14086 Default value is @file{transforms.trf}.
14089 Set how shaky the video is and how quick the camera is. It accepts an
14090 integer in the range 1-10, a value of 1 means little shakiness, a
14091 value of 10 means strong shakiness. Default value is 5.
14094 Set the accuracy of the detection process. It must be a value in the
14095 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14096 accuracy. Default value is 15.
14099 Set stepsize of the search process. The region around minimum is
14100 scanned with 1 pixel resolution. Default value is 6.
14103 Set minimum contrast. Below this value a local measurement field is
14104 discarded. Must be a floating point value in the range 0-1. Default
14108 Set reference frame number for tripod mode.
14110 If enabled, the motion of the frames is compared to a reference frame
14111 in the filtered stream, identified by the specified number. The idea
14112 is to compensate all movements in a more-or-less static scene and keep
14113 the camera view absolutely still.
14115 If set to 0, it is disabled. The frames are counted starting from 1.
14118 Show fields and transforms in the resulting frames. It accepts an
14119 integer in the range 0-2. Default value is 0, which disables any
14123 @subsection Examples
14127 Use default values:
14133 Analyze strongly shaky movie and put the results in file
14134 @file{mytransforms.trf}:
14136 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14140 Visualize the result of internal transformations in the resulting
14143 vidstabdetect=show=1
14147 Analyze a video with medium shakiness using @command{ffmpeg}:
14149 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14153 @anchor{vidstabtransform}
14154 @section vidstabtransform
14156 Video stabilization/deshaking: pass 2 of 2,
14157 see @ref{vidstabdetect} for pass 1.
14159 Read a file with transform information for each frame and
14160 apply/compensate them. Together with the @ref{vidstabdetect}
14161 filter this can be used to deshake videos. See also
14162 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14163 the @ref{unsharp} filter, see below.
14165 To enable compilation of this filter you need to configure FFmpeg with
14166 @code{--enable-libvidstab}.
14168 @subsection Options
14172 Set path to the file used to read the transforms. Default value is
14173 @file{transforms.trf}.
14176 Set the number of frames (value*2 + 1) used for lowpass filtering the
14177 camera movements. Default value is 10.
14179 For example a number of 10 means that 21 frames are used (10 in the
14180 past and 10 in the future) to smoothen the motion in the video. A
14181 larger value leads to a smoother video, but limits the acceleration of
14182 the camera (pan/tilt movements). 0 is a special case where a static
14183 camera is simulated.
14186 Set the camera path optimization algorithm.
14188 Accepted values are:
14191 gaussian kernel low-pass filter on camera motion (default)
14193 averaging on transformations
14197 Set maximal number of pixels to translate frames. Default value is -1,
14201 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14202 value is -1, meaning no limit.
14205 Specify how to deal with borders that may be visible due to movement
14208 Available values are:
14211 keep image information from previous frame (default)
14213 fill the border black
14217 Invert transforms if set to 1. Default value is 0.
14220 Consider transforms as relative to previous frame if set to 1,
14221 absolute if set to 0. Default value is 0.
14224 Set percentage to zoom. A positive value will result in a zoom-in
14225 effect, a negative value in a zoom-out effect. Default value is 0 (no
14229 Set optimal zooming to avoid borders.
14231 Accepted values are:
14236 optimal static zoom value is determined (only very strong movements
14237 will lead to visible borders) (default)
14239 optimal adaptive zoom value is determined (no borders will be
14240 visible), see @option{zoomspeed}
14243 Note that the value given at zoom is added to the one calculated here.
14246 Set percent to zoom maximally each frame (enabled when
14247 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14251 Specify type of interpolation.
14253 Available values are:
14258 linear only horizontal
14260 linear in both directions (default)
14262 cubic in both directions (slow)
14266 Enable virtual tripod mode if set to 1, which is equivalent to
14267 @code{relative=0:smoothing=0}. Default value is 0.
14269 Use also @code{tripod} option of @ref{vidstabdetect}.
14272 Increase log verbosity if set to 1. Also the detected global motions
14273 are written to the temporary file @file{global_motions.trf}. Default
14277 @subsection Examples
14281 Use @command{ffmpeg} for a typical stabilization with default values:
14283 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14286 Note the use of the @ref{unsharp} filter which is always recommended.
14289 Zoom in a bit more and load transform data from a given file:
14291 vidstabtransform=zoom=5:input="mytransforms.trf"
14295 Smoothen the video even more:
14297 vidstabtransform=smoothing=30
14303 Flip the input video vertically.
14305 For example, to vertically flip a video with @command{ffmpeg}:
14307 ffmpeg -i in.avi -vf "vflip" out.avi
14313 Make or reverse a natural vignetting effect.
14315 The filter accepts the following options:
14319 Set lens angle expression as a number of radians.
14321 The value is clipped in the @code{[0,PI/2]} range.
14323 Default value: @code{"PI/5"}
14327 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14331 Set forward/backward mode.
14333 Available modes are:
14336 The larger the distance from the central point, the darker the image becomes.
14339 The larger the distance from the central point, the brighter the image becomes.
14340 This can be used to reverse a vignette effect, though there is no automatic
14341 detection to extract the lens @option{angle} and other settings (yet). It can
14342 also be used to create a burning effect.
14345 Default value is @samp{forward}.
14348 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14350 It accepts the following values:
14353 Evaluate expressions only once during the filter initialization.
14356 Evaluate expressions for each incoming frame. This is way slower than the
14357 @samp{init} mode since it requires all the scalers to be re-computed, but it
14358 allows advanced dynamic expressions.
14361 Default value is @samp{init}.
14364 Set dithering to reduce the circular banding effects. Default is @code{1}
14368 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14369 Setting this value to the SAR of the input will make a rectangular vignetting
14370 following the dimensions of the video.
14372 Default is @code{1/1}.
14375 @subsection Expressions
14377 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14378 following parameters.
14383 input width and height
14386 the number of input frame, starting from 0
14389 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14390 @var{TB} units, NAN if undefined
14393 frame rate of the input video, NAN if the input frame rate is unknown
14396 the PTS (Presentation TimeStamp) of the filtered video frame,
14397 expressed in seconds, NAN if undefined
14400 time base of the input video
14404 @subsection Examples
14408 Apply simple strong vignetting effect:
14414 Make a flickering vignetting:
14416 vignette='PI/4+random(1)*PI/50':eval=frame
14422 Stack input videos vertically.
14424 All streams must be of same pixel format and of same width.
14426 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14427 to create same output.
14429 The filter accept the following option:
14433 Set number of input streams. Default is 2.
14436 If set to 1, force the output to terminate when the shortest input
14437 terminates. Default value is 0.
14442 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14443 Deinterlacing Filter").
14445 Based on the process described by Martin Weston for BBC R&D, and
14446 implemented based on the de-interlace algorithm written by Jim
14447 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14448 uses filter coefficients calculated by BBC R&D.
14450 There are two sets of filter coefficients, so called "simple":
14451 and "complex". Which set of filter coefficients is used can
14452 be set by passing an optional parameter:
14456 Set the interlacing filter coefficients. Accepts one of the following values:
14460 Simple filter coefficient set.
14462 More-complex filter coefficient set.
14464 Default value is @samp{complex}.
14467 Specify which frames to deinterlace. Accept one of the following values:
14471 Deinterlace all frames,
14473 Only deinterlace frames marked as interlaced.
14476 Default value is @samp{all}.
14480 Video waveform monitor.
14482 The waveform monitor plots color component intensity. By default luminance
14483 only. Each column of the waveform corresponds to a column of pixels in the
14486 It accepts the following options:
14490 Can be either @code{row}, or @code{column}. Default is @code{column}.
14491 In row mode, the graph on the left side represents color component value 0 and
14492 the right side represents value = 255. In column mode, the top side represents
14493 color component value = 0 and bottom side represents value = 255.
14496 Set intensity. Smaller values are useful to find out how many values of the same
14497 luminance are distributed across input rows/columns.
14498 Default value is @code{0.04}. Allowed range is [0, 1].
14501 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14502 In mirrored mode, higher values will be represented on the left
14503 side for @code{row} mode and at the top for @code{column} mode. Default is
14504 @code{1} (mirrored).
14508 It accepts the following values:
14511 Presents information identical to that in the @code{parade}, except
14512 that the graphs representing color components are superimposed directly
14515 This display mode makes it easier to spot relative differences or similarities
14516 in overlapping areas of the color components that are supposed to be identical,
14517 such as neutral whites, grays, or blacks.
14520 Display separate graph for the color components side by side in
14521 @code{row} mode or one below the other in @code{column} mode.
14524 Display separate graph for the color components side by side in
14525 @code{column} mode or one below the other in @code{row} mode.
14527 Using this display mode makes it easy to spot color casts in the highlights
14528 and shadows of an image, by comparing the contours of the top and the bottom
14529 graphs of each waveform. Since whites, grays, and blacks are characterized
14530 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14531 should display three waveforms of roughly equal width/height. If not, the
14532 correction is easy to perform by making level adjustments the three waveforms.
14534 Default is @code{stack}.
14536 @item components, c
14537 Set which color components to display. Default is 1, which means only luminance
14538 or red color component if input is in RGB colorspace. If is set for example to
14539 7 it will display all 3 (if) available color components.
14544 No envelope, this is default.
14547 Instant envelope, minimum and maximum values presented in graph will be easily
14548 visible even with small @code{step} value.
14551 Hold minimum and maximum values presented in graph across time. This way you
14552 can still spot out of range values without constantly looking at waveforms.
14555 Peak and instant envelope combined together.
14561 No filtering, this is default.
14564 Luma and chroma combined together.
14567 Similar as above, but shows difference between blue and red chroma.
14570 Displays only chroma.
14573 Displays actual color value on waveform.
14576 Similar as above, but with luma showing frequency of chroma values.
14580 Set which graticule to display.
14584 Do not display graticule.
14587 Display green graticule showing legal broadcast ranges.
14591 Set graticule opacity.
14594 Set graticule flags.
14598 Draw numbers above lines. By default enabled.
14601 Draw dots instead of lines.
14605 Set scale used for displaying graticule.
14612 Default is digital.
14615 Set background opacity.
14620 The @code{weave} takes a field-based video input and join
14621 each two sequential fields into single frame, producing a new double
14622 height clip with half the frame rate and half the frame count.
14624 It accepts the following option:
14628 Set first field. Available values are:
14632 Set the frame as top-field-first.
14635 Set the frame as bottom-field-first.
14639 @subsection Examples
14643 Interlace video using @ref{select} and @ref{separatefields} filter:
14645 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14650 Apply the xBR high-quality magnification filter which is designed for pixel
14651 art. It follows a set of edge-detection rules, see
14652 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14654 It accepts the following option:
14658 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14659 @code{3xBR} and @code{4} for @code{4xBR}.
14660 Default is @code{3}.
14666 Deinterlace the input video ("yadif" means "yet another deinterlacing
14669 It accepts the following parameters:
14675 The interlacing mode to adopt. It accepts one of the following values:
14678 @item 0, send_frame
14679 Output one frame for each frame.
14680 @item 1, send_field
14681 Output one frame for each field.
14682 @item 2, send_frame_nospatial
14683 Like @code{send_frame}, but it skips the spatial interlacing check.
14684 @item 3, send_field_nospatial
14685 Like @code{send_field}, but it skips the spatial interlacing check.
14688 The default value is @code{send_frame}.
14691 The picture field parity assumed for the input interlaced video. It accepts one
14692 of the following values:
14696 Assume the top field is first.
14698 Assume the bottom field is first.
14700 Enable automatic detection of field parity.
14703 The default value is @code{auto}.
14704 If the interlacing is unknown or the decoder does not export this information,
14705 top field first will be assumed.
14708 Specify which frames to deinterlace. Accept one of the following
14713 Deinterlace all frames.
14714 @item 1, interlaced
14715 Only deinterlace frames marked as interlaced.
14718 The default value is @code{all}.
14723 Apply Zoom & Pan effect.
14725 This filter accepts the following options:
14729 Set the zoom expression. Default is 1.
14733 Set the x and y expression. Default is 0.
14736 Set the duration expression in number of frames.
14737 This sets for how many number of frames effect will last for
14738 single input image.
14741 Set the output image size, default is 'hd720'.
14744 Set the output frame rate, default is '25'.
14747 Each expression can contain the following constants:
14766 Output frame count.
14770 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14771 for current input frame.
14775 'x' and 'y' of last output frame of previous input frame or 0 when there was
14776 not yet such frame (first input frame).
14779 Last calculated zoom from 'z' expression for current input frame.
14782 Last calculated zoom of last output frame of previous input frame.
14785 Number of output frames for current input frame. Calculated from 'd' expression
14786 for each input frame.
14789 number of output frames created for previous input frame
14792 Rational number: input width / input height
14795 sample aspect ratio
14798 display aspect ratio
14802 @subsection Examples
14806 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14808 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
14812 Zoom-in up to 1.5 and pan always at center of picture:
14814 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14818 Same as above but without pausing:
14820 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14825 Scale (resize) the input video, using the z.lib library:
14826 https://github.com/sekrit-twc/zimg.
14828 The zscale filter forces the output display aspect ratio to be the same
14829 as the input, by changing the output sample aspect ratio.
14831 If the input image format is different from the format requested by
14832 the next filter, the zscale filter will convert the input to the
14835 @subsection Options
14836 The filter accepts the following options.
14841 Set the output video dimension expression. Default value is the input
14844 If the @var{width} or @var{w} is 0, the input width is used for the output.
14845 If the @var{height} or @var{h} is 0, the input height is used for the output.
14847 If one of the values is -1, the zscale filter will use a value that
14848 maintains the aspect ratio of the input image, calculated from the
14849 other specified dimension. If both of them are -1, the input size is
14852 If one of the values is -n with n > 1, the zscale filter will also use a value
14853 that maintains the aspect ratio of the input image, calculated from the other
14854 specified dimension. After that it will, however, make sure that the calculated
14855 dimension is divisible by n and adjust the value if necessary.
14857 See below for the list of accepted constants for use in the dimension
14861 Set the video size. For the syntax of this option, check the
14862 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14865 Set the dither type.
14867 Possible values are:
14872 @item error_diffusion
14878 Set the resize filter type.
14880 Possible values are:
14890 Default is bilinear.
14893 Set the color range.
14895 Possible values are:
14902 Default is same as input.
14905 Set the color primaries.
14907 Possible values are:
14917 Default is same as input.
14920 Set the transfer characteristics.
14922 Possible values are:
14936 Default is same as input.
14939 Set the colorspace matrix.
14941 Possible value are:
14952 Default is same as input.
14955 Set the input color range.
14957 Possible values are:
14964 Default is same as input.
14966 @item primariesin, pin
14967 Set the input color primaries.
14969 Possible values are:
14979 Default is same as input.
14981 @item transferin, tin
14982 Set the input transfer characteristics.
14984 Possible values are:
14995 Default is same as input.
14997 @item matrixin, min
14998 Set the input colorspace matrix.
15000 Possible value are:
15012 Set the output chroma location.
15014 Possible values are:
15025 @item chromalin, cin
15026 Set the input chroma location.
15028 Possible values are:
15040 Set the nominal peak luminance.
15043 The values of the @option{w} and @option{h} options are expressions
15044 containing the following constants:
15049 The input width and height
15053 These are the same as @var{in_w} and @var{in_h}.
15057 The output (scaled) width and height
15061 These are the same as @var{out_w} and @var{out_h}
15064 The same as @var{iw} / @var{ih}
15067 input sample aspect ratio
15070 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15074 horizontal and vertical input chroma subsample values. For example for the
15075 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15079 horizontal and vertical output chroma subsample values. For example for the
15080 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15086 @c man end VIDEO FILTERS
15088 @chapter Video Sources
15089 @c man begin VIDEO SOURCES
15091 Below is a description of the currently available video sources.
15095 Buffer video frames, and make them available to the filter chain.
15097 This source is mainly intended for a programmatic use, in particular
15098 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15100 It accepts the following parameters:
15105 Specify the size (width and height) of the buffered video frames. For the
15106 syntax of this option, check the
15107 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15110 The input video width.
15113 The input video height.
15116 A string representing the pixel format of the buffered video frames.
15117 It may be a number corresponding to a pixel format, or a pixel format
15121 Specify the timebase assumed by the timestamps of the buffered frames.
15124 Specify the frame rate expected for the video stream.
15126 @item pixel_aspect, sar
15127 The sample (pixel) aspect ratio of the input video.
15130 Specify the optional parameters to be used for the scale filter which
15131 is automatically inserted when an input change is detected in the
15132 input size or format.
15134 @item hw_frames_ctx
15135 When using a hardware pixel format, this should be a reference to an
15136 AVHWFramesContext describing input frames.
15141 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15144 will instruct the source to accept video frames with size 320x240 and
15145 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15146 square pixels (1:1 sample aspect ratio).
15147 Since the pixel format with name "yuv410p" corresponds to the number 6
15148 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15149 this example corresponds to:
15151 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15154 Alternatively, the options can be specified as a flat string, but this
15155 syntax is deprecated:
15157 @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}]
15161 Create a pattern generated by an elementary cellular automaton.
15163 The initial state of the cellular automaton can be defined through the
15164 @option{filename} and @option{pattern} options. If such options are
15165 not specified an initial state is created randomly.
15167 At each new frame a new row in the video is filled with the result of
15168 the cellular automaton next generation. The behavior when the whole
15169 frame is filled is defined by the @option{scroll} option.
15171 This source accepts the following options:
15175 Read the initial cellular automaton state, i.e. the starting row, from
15176 the specified file.
15177 In the file, each non-whitespace character is considered an alive
15178 cell, a newline will terminate the row, and further characters in the
15179 file will be ignored.
15182 Read the initial cellular automaton state, i.e. the starting row, from
15183 the specified string.
15185 Each non-whitespace character in the string is considered an alive
15186 cell, a newline will terminate the row, and further characters in the
15187 string will be ignored.
15190 Set the video rate, that is the number of frames generated per second.
15193 @item random_fill_ratio, ratio
15194 Set the random fill ratio for the initial cellular automaton row. It
15195 is a floating point number value ranging from 0 to 1, defaults to
15198 This option is ignored when a file or a pattern is specified.
15200 @item random_seed, seed
15201 Set the seed for filling randomly the initial row, must be an integer
15202 included between 0 and UINT32_MAX. If not specified, or if explicitly
15203 set to -1, the filter will try to use a good random seed on a best
15207 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15208 Default value is 110.
15211 Set the size of the output video. For the syntax of this option, check the
15212 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15214 If @option{filename} or @option{pattern} is specified, the size is set
15215 by default to the width of the specified initial state row, and the
15216 height is set to @var{width} * PHI.
15218 If @option{size} is set, it must contain the width of the specified
15219 pattern string, and the specified pattern will be centered in the
15222 If a filename or a pattern string is not specified, the size value
15223 defaults to "320x518" (used for a randomly generated initial state).
15226 If set to 1, scroll the output upward when all the rows in the output
15227 have been already filled. If set to 0, the new generated row will be
15228 written over the top row just after the bottom row is filled.
15231 @item start_full, full
15232 If set to 1, completely fill the output with generated rows before
15233 outputting the first frame.
15234 This is the default behavior, for disabling set the value to 0.
15237 If set to 1, stitch the left and right row edges together.
15238 This is the default behavior, for disabling set the value to 0.
15241 @subsection Examples
15245 Read the initial state from @file{pattern}, and specify an output of
15248 cellauto=f=pattern:s=200x400
15252 Generate a random initial row with a width of 200 cells, with a fill
15255 cellauto=ratio=2/3:s=200x200
15259 Create a pattern generated by rule 18 starting by a single alive cell
15260 centered on an initial row with width 100:
15262 cellauto=p=@@:s=100x400:full=0:rule=18
15266 Specify a more elaborated initial pattern:
15268 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15273 @anchor{coreimagesrc}
15274 @section coreimagesrc
15275 Video source generated on GPU using Apple's CoreImage API on OSX.
15277 This video source is a specialized version of the @ref{coreimage} video filter.
15278 Use a core image generator at the beginning of the applied filterchain to
15279 generate the content.
15281 The coreimagesrc video source accepts the following options:
15283 @item list_generators
15284 List all available generators along with all their respective options as well as
15285 possible minimum and maximum values along with the default values.
15287 list_generators=true
15291 Specify the size of the sourced video. For the syntax of this option, check the
15292 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15293 The default value is @code{320x240}.
15296 Specify the frame rate of the sourced video, as the number of frames
15297 generated per second. It has to be a string in the format
15298 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15299 number or a valid video frame rate abbreviation. The default value is
15303 Set the sample aspect ratio of the sourced video.
15306 Set the duration of the sourced video. See
15307 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15308 for the accepted syntax.
15310 If not specified, or the expressed duration is negative, the video is
15311 supposed to be generated forever.
15314 Additionally, all options of the @ref{coreimage} video filter are accepted.
15315 A complete filterchain can be used for further processing of the
15316 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15317 and examples for details.
15319 @subsection Examples
15324 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15325 given as complete and escaped command-line for Apple's standard bash shell:
15327 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15329 This example is equivalent to the QRCode example of @ref{coreimage} without the
15330 need for a nullsrc video source.
15334 @section mandelbrot
15336 Generate a Mandelbrot set fractal, and progressively zoom towards the
15337 point specified with @var{start_x} and @var{start_y}.
15339 This source accepts the following options:
15344 Set the terminal pts value. Default value is 400.
15347 Set the terminal scale value.
15348 Must be a floating point value. Default value is 0.3.
15351 Set the inner coloring mode, that is the algorithm used to draw the
15352 Mandelbrot fractal internal region.
15354 It shall assume one of the following values:
15359 Show time until convergence.
15361 Set color based on point closest to the origin of the iterations.
15366 Default value is @var{mincol}.
15369 Set the bailout value. Default value is 10.0.
15372 Set the maximum of iterations performed by the rendering
15373 algorithm. Default value is 7189.
15376 Set outer coloring mode.
15377 It shall assume one of following values:
15379 @item iteration_count
15380 Set iteration cound mode.
15381 @item normalized_iteration_count
15382 set normalized iteration count mode.
15384 Default value is @var{normalized_iteration_count}.
15387 Set frame rate, expressed as number of frames per second. Default
15391 Set frame size. For the syntax of this option, check the "Video
15392 size" section in the ffmpeg-utils manual. Default value is "640x480".
15395 Set the initial scale value. Default value is 3.0.
15398 Set the initial x position. Must be a floating point value between
15399 -100 and 100. Default value is -0.743643887037158704752191506114774.
15402 Set the initial y position. Must be a floating point value between
15403 -100 and 100. Default value is -0.131825904205311970493132056385139.
15408 Generate various test patterns, as generated by the MPlayer test filter.
15410 The size of the generated video is fixed, and is 256x256.
15411 This source is useful in particular for testing encoding features.
15413 This source accepts the following options:
15418 Specify the frame rate of the sourced video, as the number of frames
15419 generated per second. It has to be a string in the format
15420 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15421 number or a valid video frame rate abbreviation. The default value is
15425 Set the duration of the sourced video. See
15426 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15427 for the accepted syntax.
15429 If not specified, or the expressed duration is negative, the video is
15430 supposed to be generated forever.
15434 Set the number or the name of the test to perform. Supported tests are:
15450 Default value is "all", which will cycle through the list of all tests.
15455 mptestsrc=t=dc_luma
15458 will generate a "dc_luma" test pattern.
15460 @section frei0r_src
15462 Provide a frei0r source.
15464 To enable compilation of this filter you need to install the frei0r
15465 header and configure FFmpeg with @code{--enable-frei0r}.
15467 This source accepts the following parameters:
15472 The size of the video to generate. For the syntax of this option, check the
15473 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15476 The framerate of the generated video. It may be a string of the form
15477 @var{num}/@var{den} or a frame rate abbreviation.
15480 The name to the frei0r source to load. For more information regarding frei0r and
15481 how to set the parameters, read the @ref{frei0r} section in the video filters
15484 @item filter_params
15485 A '|'-separated list of parameters to pass to the frei0r source.
15489 For example, to generate a frei0r partik0l source with size 200x200
15490 and frame rate 10 which is overlaid on the overlay filter main input:
15492 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15497 Generate a life pattern.
15499 This source is based on a generalization of John Conway's life game.
15501 The sourced input represents a life grid, each pixel represents a cell
15502 which can be in one of two possible states, alive or dead. Every cell
15503 interacts with its eight neighbours, which are the cells that are
15504 horizontally, vertically, or diagonally adjacent.
15506 At each interaction the grid evolves according to the adopted rule,
15507 which specifies the number of neighbor alive cells which will make a
15508 cell stay alive or born. The @option{rule} option allows one to specify
15511 This source accepts the following options:
15515 Set the file from which to read the initial grid state. In the file,
15516 each non-whitespace character is considered an alive cell, and newline
15517 is used to delimit the end of each row.
15519 If this option is not specified, the initial grid is generated
15523 Set the video rate, that is the number of frames generated per second.
15526 @item random_fill_ratio, ratio
15527 Set the random fill ratio for the initial random grid. It is a
15528 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15529 It is ignored when a file is specified.
15531 @item random_seed, seed
15532 Set the seed for filling the initial random grid, must be an integer
15533 included between 0 and UINT32_MAX. If not specified, or if explicitly
15534 set to -1, the filter will try to use a good random seed on a best
15540 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15541 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15542 @var{NS} specifies the number of alive neighbor cells which make a
15543 live cell stay alive, and @var{NB} the number of alive neighbor cells
15544 which make a dead cell to become alive (i.e. to "born").
15545 "s" and "b" can be used in place of "S" and "B", respectively.
15547 Alternatively a rule can be specified by an 18-bits integer. The 9
15548 high order bits are used to encode the next cell state if it is alive
15549 for each number of neighbor alive cells, the low order bits specify
15550 the rule for "borning" new cells. Higher order bits encode for an
15551 higher number of neighbor cells.
15552 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15553 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15555 Default value is "S23/B3", which is the original Conway's game of life
15556 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15557 cells, and will born a new cell if there are three alive cells around
15561 Set the size of the output video. For the syntax of this option, check the
15562 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15564 If @option{filename} is specified, the size is set by default to the
15565 same size of the input file. If @option{size} is set, it must contain
15566 the size specified in the input file, and the initial grid defined in
15567 that file is centered in the larger resulting area.
15569 If a filename is not specified, the size value defaults to "320x240"
15570 (used for a randomly generated initial grid).
15573 If set to 1, stitch the left and right grid edges together, and the
15574 top and bottom edges also. Defaults to 1.
15577 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15578 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15579 value from 0 to 255.
15582 Set the color of living (or new born) cells.
15585 Set the color of dead cells. If @option{mold} is set, this is the first color
15586 used to represent a dead cell.
15589 Set mold color, for definitely dead and moldy cells.
15591 For the syntax of these 3 color options, check the "Color" section in the
15592 ffmpeg-utils manual.
15595 @subsection Examples
15599 Read a grid from @file{pattern}, and center it on a grid of size
15602 life=f=pattern:s=300x300
15606 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15608 life=ratio=2/3:s=200x200
15612 Specify a custom rule for evolving a randomly generated grid:
15618 Full example with slow death effect (mold) using @command{ffplay}:
15620 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15627 @anchor{haldclutsrc}
15629 @anchor{rgbtestsrc}
15631 @anchor{smptehdbars}
15634 @anchor{yuvtestsrc}
15635 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15637 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15639 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15641 The @code{color} source provides an uniformly colored input.
15643 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15644 @ref{haldclut} filter.
15646 The @code{nullsrc} source returns unprocessed video frames. It is
15647 mainly useful to be employed in analysis / debugging tools, or as the
15648 source for filters which ignore the input data.
15650 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15651 detecting RGB vs BGR issues. You should see a red, green and blue
15652 stripe from top to bottom.
15654 The @code{smptebars} source generates a color bars pattern, based on
15655 the SMPTE Engineering Guideline EG 1-1990.
15657 The @code{smptehdbars} source generates a color bars pattern, based on
15658 the SMPTE RP 219-2002.
15660 The @code{testsrc} source generates a test video pattern, showing a
15661 color pattern, a scrolling gradient and a timestamp. This is mainly
15662 intended for testing purposes.
15664 The @code{testsrc2} source is similar to testsrc, but supports more
15665 pixel formats instead of just @code{rgb24}. This allows using it as an
15666 input for other tests without requiring a format conversion.
15668 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15669 see a y, cb and cr stripe from top to bottom.
15671 The sources accept the following parameters:
15676 Specify the color of the source, only available in the @code{color}
15677 source. For the syntax of this option, check the "Color" section in the
15678 ffmpeg-utils manual.
15681 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15682 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15683 pixels to be used as identity matrix for 3D lookup tables. Each component is
15684 coded on a @code{1/(N*N)} scale.
15687 Specify the size of the sourced video. For the syntax of this option, check the
15688 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15689 The default value is @code{320x240}.
15691 This option is not available with the @code{haldclutsrc} filter.
15694 Specify the frame rate of the sourced video, as the number of frames
15695 generated per second. It has to be a string in the format
15696 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15697 number or a valid video frame rate abbreviation. The default value is
15701 Set the sample aspect ratio of the sourced video.
15704 Set the duration of the sourced video. See
15705 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15706 for the accepted syntax.
15708 If not specified, or the expressed duration is negative, the video is
15709 supposed to be generated forever.
15712 Set the number of decimals to show in the timestamp, only available in the
15713 @code{testsrc} source.
15715 The displayed timestamp value will correspond to the original
15716 timestamp value multiplied by the power of 10 of the specified
15717 value. Default value is 0.
15720 For example the following:
15722 testsrc=duration=5.3:size=qcif:rate=10
15725 will generate a video with a duration of 5.3 seconds, with size
15726 176x144 and a frame rate of 10 frames per second.
15728 The following graph description will generate a red source
15729 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15732 color=c=red@@0.2:s=qcif:r=10
15735 If the input content is to be ignored, @code{nullsrc} can be used. The
15736 following command generates noise in the luminance plane by employing
15737 the @code{geq} filter:
15739 nullsrc=s=256x256, geq=random(1)*255:128:128
15742 @subsection Commands
15744 The @code{color} source supports the following commands:
15748 Set the color of the created image. Accepts the same syntax of the
15749 corresponding @option{color} option.
15752 @c man end VIDEO SOURCES
15754 @chapter Video Sinks
15755 @c man begin VIDEO SINKS
15757 Below is a description of the currently available video sinks.
15759 @section buffersink
15761 Buffer video frames, and make them available to the end of the filter
15764 This sink is mainly intended for programmatic use, in particular
15765 through the interface defined in @file{libavfilter/buffersink.h}
15766 or the options system.
15768 It accepts a pointer to an AVBufferSinkContext structure, which
15769 defines the incoming buffers' formats, to be passed as the opaque
15770 parameter to @code{avfilter_init_filter} for initialization.
15774 Null video sink: do absolutely nothing with the input video. It is
15775 mainly useful as a template and for use in analysis / debugging
15778 @c man end VIDEO SINKS
15780 @chapter Multimedia Filters
15781 @c man begin MULTIMEDIA FILTERS
15783 Below is a description of the currently available multimedia filters.
15787 Convert input audio to a video output, displaying the audio bit scope.
15789 The filter accepts the following options:
15793 Set frame rate, expressed as number of frames per second. Default
15797 Specify the video size for the output. For the syntax of this option, check the
15798 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15799 Default value is @code{1024x256}.
15802 Specify list of colors separated by space or by '|' which will be used to
15803 draw channels. Unrecognized or missing colors will be replaced
15807 @section ahistogram
15809 Convert input audio to a video output, displaying the volume histogram.
15811 The filter accepts the following options:
15815 Specify how histogram is calculated.
15817 It accepts the following values:
15820 Use single histogram for all channels.
15822 Use separate histogram for each channel.
15824 Default is @code{single}.
15827 Set frame rate, expressed as number of frames per second. Default
15831 Specify the video size for the output. For the syntax of this option, check the
15832 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15833 Default value is @code{hd720}.
15838 It accepts the following values:
15849 reverse logarithmic
15851 Default is @code{log}.
15854 Set amplitude scale.
15856 It accepts the following values:
15863 Default is @code{log}.
15866 Set how much frames to accumulate in histogram.
15867 Defauls is 1. Setting this to -1 accumulates all frames.
15870 Set histogram ratio of window height.
15873 Set sonogram sliding.
15875 It accepts the following values:
15878 replace old rows with new ones.
15880 scroll from top to bottom.
15882 Default is @code{replace}.
15885 @section aphasemeter
15887 Convert input audio to a video output, displaying the audio phase.
15889 The filter accepts the following options:
15893 Set the output frame rate. Default value is @code{25}.
15896 Set the video size for the output. For the syntax of this option, check the
15897 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15898 Default value is @code{800x400}.
15903 Specify the red, green, blue contrast. Default values are @code{2},
15904 @code{7} and @code{1}.
15905 Allowed range is @code{[0, 255]}.
15908 Set color which will be used for drawing median phase. If color is
15909 @code{none} which is default, no median phase value will be drawn.
15912 Enable video output. Default is enabled.
15915 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
15916 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
15917 The @code{-1} means left and right channels are completely out of phase and
15918 @code{1} means channels are in phase.
15920 @section avectorscope
15922 Convert input audio to a video output, representing the audio vector
15925 The filter is used to measure the difference between channels of stereo
15926 audio stream. A monoaural signal, consisting of identical left and right
15927 signal, results in straight vertical line. Any stereo separation is visible
15928 as a deviation from this line, creating a Lissajous figure.
15929 If the straight (or deviation from it) but horizontal line appears this
15930 indicates that the left and right channels are out of phase.
15932 The filter accepts the following options:
15936 Set the vectorscope mode.
15938 Available values are:
15941 Lissajous rotated by 45 degrees.
15944 Same as above but not rotated.
15947 Shape resembling half of circle.
15950 Default value is @samp{lissajous}.
15953 Set the video size for the output. For the syntax of this option, check the
15954 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15955 Default value is @code{400x400}.
15958 Set the output frame rate. Default value is @code{25}.
15964 Specify the red, green, blue and alpha contrast. Default values are @code{40},
15965 @code{160}, @code{80} and @code{255}.
15966 Allowed range is @code{[0, 255]}.
15972 Specify the red, green, blue and alpha fade. Default values are @code{15},
15973 @code{10}, @code{5} and @code{5}.
15974 Allowed range is @code{[0, 255]}.
15977 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
15980 Set the vectorscope drawing mode.
15982 Available values are:
15985 Draw dot for each sample.
15988 Draw line between previous and current sample.
15991 Default value is @samp{dot}.
15994 Specify amplitude scale of audio samples.
15996 Available values are:
16013 @subsection Examples
16017 Complete example using @command{ffplay}:
16019 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16020 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16024 @section bench, abench
16026 Benchmark part of a filtergraph.
16028 The filter accepts the following options:
16032 Start or stop a timer.
16034 Available values are:
16037 Get the current time, set it as frame metadata (using the key
16038 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
16041 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16042 the input frame metadata to get the time difference. Time difference, average,
16043 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16044 @code{min}) are then printed. The timestamps are expressed in seconds.
16048 @subsection Examples
16052 Benchmark @ref{selectivecolor} filter:
16054 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16060 Concatenate audio and video streams, joining them together one after the
16063 The filter works on segments of synchronized video and audio streams. All
16064 segments must have the same number of streams of each type, and that will
16065 also be the number of streams at output.
16067 The filter accepts the following options:
16072 Set the number of segments. Default is 2.
16075 Set the number of output video streams, that is also the number of video
16076 streams in each segment. Default is 1.
16079 Set the number of output audio streams, that is also the number of audio
16080 streams in each segment. Default is 0.
16083 Activate unsafe mode: do not fail if segments have a different format.
16087 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16088 @var{a} audio outputs.
16090 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16091 segment, in the same order as the outputs, then the inputs for the second
16094 Related streams do not always have exactly the same duration, for various
16095 reasons including codec frame size or sloppy authoring. For that reason,
16096 related synchronized streams (e.g. a video and its audio track) should be
16097 concatenated at once. The concat filter will use the duration of the longest
16098 stream in each segment (except the last one), and if necessary pad shorter
16099 audio streams with silence.
16101 For this filter to work correctly, all segments must start at timestamp 0.
16103 All corresponding streams must have the same parameters in all segments; the
16104 filtering system will automatically select a common pixel format for video
16105 streams, and a common sample format, sample rate and channel layout for
16106 audio streams, but other settings, such as resolution, must be converted
16107 explicitly by the user.
16109 Different frame rates are acceptable but will result in variable frame rate
16110 at output; be sure to configure the output file to handle it.
16112 @subsection Examples
16116 Concatenate an opening, an episode and an ending, all in bilingual version
16117 (video in stream 0, audio in streams 1 and 2):
16119 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16120 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16121 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16122 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16126 Concatenate two parts, handling audio and video separately, using the
16127 (a)movie sources, and adjusting the resolution:
16129 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16130 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16131 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16133 Note that a desync will happen at the stitch if the audio and video streams
16134 do not have exactly the same duration in the first file.
16138 @section drawgraph, adrawgraph
16140 Draw a graph using input video or audio metadata.
16142 It accepts the following parameters:
16146 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16149 Set 1st foreground color expression.
16152 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16155 Set 2nd foreground color expression.
16158 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16161 Set 3rd foreground color expression.
16164 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16167 Set 4th foreground color expression.
16170 Set minimal value of metadata value.
16173 Set maximal value of metadata value.
16176 Set graph background color. Default is white.
16181 Available values for mode is:
16188 Default is @code{line}.
16193 Available values for slide is:
16196 Draw new frame when right border is reached.
16199 Replace old columns with new ones.
16202 Scroll from right to left.
16205 Scroll from left to right.
16208 Draw single picture.
16211 Default is @code{frame}.
16214 Set size of graph video. For the syntax of this option, check the
16215 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16216 The default value is @code{900x256}.
16218 The foreground color expressions can use the following variables:
16221 Minimal value of metadata value.
16224 Maximal value of metadata value.
16227 Current metadata key value.
16230 The color is defined as 0xAABBGGRR.
16233 Example using metadata from @ref{signalstats} filter:
16235 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16238 Example using metadata from @ref{ebur128} filter:
16240 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16246 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16247 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16248 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16249 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16251 The filter also has a video output (see the @var{video} option) with a real
16252 time graph to observe the loudness evolution. The graphic contains the logged
16253 message mentioned above, so it is not printed anymore when this option is set,
16254 unless the verbose logging is set. The main graphing area contains the
16255 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16256 the momentary loudness (400 milliseconds).
16258 More information about the Loudness Recommendation EBU R128 on
16259 @url{http://tech.ebu.ch/loudness}.
16261 The filter accepts the following options:
16266 Activate the video output. The audio stream is passed unchanged whether this
16267 option is set or no. The video stream will be the first output stream if
16268 activated. Default is @code{0}.
16271 Set the video size. This option is for video only. For the syntax of this
16273 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16274 Default and minimum resolution is @code{640x480}.
16277 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16278 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16279 other integer value between this range is allowed.
16282 Set metadata injection. If set to @code{1}, the audio input will be segmented
16283 into 100ms output frames, each of them containing various loudness information
16284 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16286 Default is @code{0}.
16289 Force the frame logging level.
16291 Available values are:
16294 information logging level
16296 verbose logging level
16299 By default, the logging level is set to @var{info}. If the @option{video} or
16300 the @option{metadata} options are set, it switches to @var{verbose}.
16305 Available modes can be cumulated (the option is a @code{flag} type). Possible
16309 Disable any peak mode (default).
16311 Enable sample-peak mode.
16313 Simple peak mode looking for the higher sample value. It logs a message
16314 for sample-peak (identified by @code{SPK}).
16316 Enable true-peak mode.
16318 If enabled, the peak lookup is done on an over-sampled version of the input
16319 stream for better peak accuracy. It logs a message for true-peak.
16320 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16321 This mode requires a build with @code{libswresample}.
16325 Treat mono input files as "dual mono". If a mono file is intended for playback
16326 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16327 If set to @code{true}, this option will compensate for this effect.
16328 Multi-channel input files are not affected by this option.
16331 Set a specific pan law to be used for the measurement of dual mono files.
16332 This parameter is optional, and has a default value of -3.01dB.
16335 @subsection Examples
16339 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16341 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16345 Run an analysis with @command{ffmpeg}:
16347 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16351 @section interleave, ainterleave
16353 Temporally interleave frames from several inputs.
16355 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16357 These filters read frames from several inputs and send the oldest
16358 queued frame to the output.
16360 Input streams must have well defined, monotonically increasing frame
16363 In order to submit one frame to output, these filters need to enqueue
16364 at least one frame for each input, so they cannot work in case one
16365 input is not yet terminated and will not receive incoming frames.
16367 For example consider the case when one input is a @code{select} filter
16368 which always drops input frames. The @code{interleave} filter will keep
16369 reading from that input, but it will never be able to send new frames
16370 to output until the input sends an end-of-stream signal.
16372 Also, depending on inputs synchronization, the filters will drop
16373 frames in case one input receives more frames than the other ones, and
16374 the queue is already filled.
16376 These filters accept the following options:
16380 Set the number of different inputs, it is 2 by default.
16383 @subsection Examples
16387 Interleave frames belonging to different streams using @command{ffmpeg}:
16389 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16393 Add flickering blur effect:
16395 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16399 @section metadata, ametadata
16401 Manipulate frame metadata.
16403 This filter accepts the following options:
16407 Set mode of operation of the filter.
16409 Can be one of the following:
16413 If both @code{value} and @code{key} is set, select frames
16414 which have such metadata. If only @code{key} is set, select
16415 every frame that has such key in metadata.
16418 Add new metadata @code{key} and @code{value}. If key is already available
16422 Modify value of already present key.
16425 If @code{value} is set, delete only keys that have such value.
16426 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16430 Print key and its value if metadata was found. If @code{key} is not set print all
16431 metadata values available in frame.
16435 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16438 Set metadata value which will be used. This option is mandatory for
16439 @code{modify} and @code{add} mode.
16442 Which function to use when comparing metadata value and @code{value}.
16444 Can be one of following:
16448 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16451 Values are interpreted as strings, returns true if metadata value starts with
16452 the @code{value} option string.
16455 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16458 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16461 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16464 Values are interpreted as floats, returns true if expression from option @code{expr}
16469 Set expression which is used when @code{function} is set to @code{expr}.
16470 The expression is evaluated through the eval API and can contain the following
16475 Float representation of @code{value} from metadata key.
16478 Float representation of @code{value} as supplied by user in @code{value} option.
16482 If specified in @code{print} mode, output is written to the named file. Instead of
16483 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16484 for standard output. If @code{file} option is not set, output is written to the log
16485 with AV_LOG_INFO loglevel.
16489 @subsection Examples
16493 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16496 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16499 Print silencedetect output to file @file{metadata.txt}.
16501 silencedetect,ametadata=mode=print:file=metadata.txt
16504 Direct all metadata to a pipe with file descriptor 4.
16506 metadata=mode=print:file='pipe\:4'
16510 @section perms, aperms
16512 Set read/write permissions for the output frames.
16514 These filters are mainly aimed at developers to test direct path in the
16515 following filter in the filtergraph.
16517 The filters accept the following options:
16521 Select the permissions mode.
16523 It accepts the following values:
16526 Do nothing. This is the default.
16528 Set all the output frames read-only.
16530 Set all the output frames directly writable.
16532 Make the frame read-only if writable, and writable if read-only.
16534 Set each output frame read-only or writable randomly.
16538 Set the seed for the @var{random} mode, must be an integer included between
16539 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16540 @code{-1}, the filter will try to use a good random seed on a best effort
16544 Note: in case of auto-inserted filter between the permission filter and the
16545 following one, the permission might not be received as expected in that
16546 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16547 perms/aperms filter can avoid this problem.
16549 @section realtime, arealtime
16551 Slow down filtering to match real time approximatively.
16553 These filters will pause the filtering for a variable amount of time to
16554 match the output rate with the input timestamps.
16555 They are similar to the @option{re} option to @code{ffmpeg}.
16557 They accept the following options:
16561 Time limit for the pauses. Any pause longer than that will be considered
16562 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16566 @section select, aselect
16568 Select frames to pass in output.
16570 This filter accepts the following options:
16575 Set expression, which is evaluated for each input frame.
16577 If the expression is evaluated to zero, the frame is discarded.
16579 If the evaluation result is negative or NaN, the frame is sent to the
16580 first output; otherwise it is sent to the output with index
16581 @code{ceil(val)-1}, assuming that the input index starts from 0.
16583 For example a value of @code{1.2} corresponds to the output with index
16584 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16587 Set the number of outputs. The output to which to send the selected
16588 frame is based on the result of the evaluation. Default value is 1.
16591 The expression can contain the following constants:
16595 The (sequential) number of the filtered frame, starting from 0.
16598 The (sequential) number of the selected frame, starting from 0.
16600 @item prev_selected_n
16601 The sequential number of the last selected frame. It's NAN if undefined.
16604 The timebase of the input timestamps.
16607 The PTS (Presentation TimeStamp) of the filtered video frame,
16608 expressed in @var{TB} units. It's NAN if undefined.
16611 The PTS of the filtered video frame,
16612 expressed in seconds. It's NAN if undefined.
16615 The PTS of the previously filtered video frame. It's NAN if undefined.
16617 @item prev_selected_pts
16618 The PTS of the last previously filtered video frame. It's NAN if undefined.
16620 @item prev_selected_t
16621 The PTS of the last previously selected video frame. It's NAN if undefined.
16624 The PTS of the first video frame in the video. It's NAN if undefined.
16627 The time of the first video frame in the video. It's NAN if undefined.
16629 @item pict_type @emph{(video only)}
16630 The type of the filtered frame. It can assume one of the following
16642 @item interlace_type @emph{(video only)}
16643 The frame interlace type. It can assume one of the following values:
16646 The frame is progressive (not interlaced).
16648 The frame is top-field-first.
16650 The frame is bottom-field-first.
16653 @item consumed_sample_n @emph{(audio only)}
16654 the number of selected samples before the current frame
16656 @item samples_n @emph{(audio only)}
16657 the number of samples in the current frame
16659 @item sample_rate @emph{(audio only)}
16660 the input sample rate
16663 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16666 the position in the file of the filtered frame, -1 if the information
16667 is not available (e.g. for synthetic video)
16669 @item scene @emph{(video only)}
16670 value between 0 and 1 to indicate a new scene; a low value reflects a low
16671 probability for the current frame to introduce a new scene, while a higher
16672 value means the current frame is more likely to be one (see the example below)
16674 @item concatdec_select
16675 The concat demuxer can select only part of a concat input file by setting an
16676 inpoint and an outpoint, but the output packets may not be entirely contained
16677 in the selected interval. By using this variable, it is possible to skip frames
16678 generated by the concat demuxer which are not exactly contained in the selected
16681 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16682 and the @var{lavf.concat.duration} packet metadata values which are also
16683 present in the decoded frames.
16685 The @var{concatdec_select} variable is -1 if the frame pts is at least
16686 start_time and either the duration metadata is missing or the frame pts is less
16687 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16690 That basically means that an input frame is selected if its pts is within the
16691 interval set by the concat demuxer.
16695 The default value of the select expression is "1".
16697 @subsection Examples
16701 Select all frames in input:
16706 The example above is the same as:
16718 Select only I-frames:
16720 select='eq(pict_type\,I)'
16724 Select one frame every 100:
16726 select='not(mod(n\,100))'
16730 Select only frames contained in the 10-20 time interval:
16732 select=between(t\,10\,20)
16736 Select only I-frames contained in the 10-20 time interval:
16738 select=between(t\,10\,20)*eq(pict_type\,I)
16742 Select frames with a minimum distance of 10 seconds:
16744 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16748 Use aselect to select only audio frames with samples number > 100:
16750 aselect='gt(samples_n\,100)'
16754 Create a mosaic of the first scenes:
16756 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16759 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16763 Send even and odd frames to separate outputs, and compose them:
16765 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16769 Select useful frames from an ffconcat file which is using inpoints and
16770 outpoints but where the source files are not intra frame only.
16772 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16776 @section sendcmd, asendcmd
16778 Send commands to filters in the filtergraph.
16780 These filters read commands to be sent to other filters in the
16783 @code{sendcmd} must be inserted between two video filters,
16784 @code{asendcmd} must be inserted between two audio filters, but apart
16785 from that they act the same way.
16787 The specification of commands can be provided in the filter arguments
16788 with the @var{commands} option, or in a file specified by the
16789 @var{filename} option.
16791 These filters accept the following options:
16794 Set the commands to be read and sent to the other filters.
16796 Set the filename of the commands to be read and sent to the other
16800 @subsection Commands syntax
16802 A commands description consists of a sequence of interval
16803 specifications, comprising a list of commands to be executed when a
16804 particular event related to that interval occurs. The occurring event
16805 is typically the current frame time entering or leaving a given time
16808 An interval is specified by the following syntax:
16810 @var{START}[-@var{END}] @var{COMMANDS};
16813 The time interval is specified by the @var{START} and @var{END} times.
16814 @var{END} is optional and defaults to the maximum time.
16816 The current frame time is considered within the specified interval if
16817 it is included in the interval [@var{START}, @var{END}), that is when
16818 the time is greater or equal to @var{START} and is lesser than
16821 @var{COMMANDS} consists of a sequence of one or more command
16822 specifications, separated by ",", relating to that interval. The
16823 syntax of a command specification is given by:
16825 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16828 @var{FLAGS} is optional and specifies the type of events relating to
16829 the time interval which enable sending the specified command, and must
16830 be a non-null sequence of identifier flags separated by "+" or "|" and
16831 enclosed between "[" and "]".
16833 The following flags are recognized:
16836 The command is sent when the current frame timestamp enters the
16837 specified interval. In other words, the command is sent when the
16838 previous frame timestamp was not in the given interval, and the
16842 The command is sent when the current frame timestamp leaves the
16843 specified interval. In other words, the command is sent when the
16844 previous frame timestamp was in the given interval, and the
16848 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16851 @var{TARGET} specifies the target of the command, usually the name of
16852 the filter class or a specific filter instance name.
16854 @var{COMMAND} specifies the name of the command for the target filter.
16856 @var{ARG} is optional and specifies the optional list of argument for
16857 the given @var{COMMAND}.
16859 Between one interval specification and another, whitespaces, or
16860 sequences of characters starting with @code{#} until the end of line,
16861 are ignored and can be used to annotate comments.
16863 A simplified BNF description of the commands specification syntax
16866 @var{COMMAND_FLAG} ::= "enter" | "leave"
16867 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16868 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16869 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16870 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16871 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16874 @subsection Examples
16878 Specify audio tempo change at second 4:
16880 asendcmd=c='4.0 atempo tempo 1.5',atempo
16884 Specify a list of drawtext and hue commands in a file.
16886 # show text in the interval 5-10
16887 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16888 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16890 # desaturate the image in the interval 15-20
16891 15.0-20.0 [enter] hue s 0,
16892 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16894 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16896 # apply an exponential saturation fade-out effect, starting from time 25
16897 25 [enter] hue s exp(25-t)
16900 A filtergraph allowing to read and process the above command list
16901 stored in a file @file{test.cmd}, can be specified with:
16903 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
16908 @section setpts, asetpts
16910 Change the PTS (presentation timestamp) of the input frames.
16912 @code{setpts} works on video frames, @code{asetpts} on audio frames.
16914 This filter accepts the following options:
16919 The expression which is evaluated for each frame to construct its timestamp.
16923 The expression is evaluated through the eval API and can contain the following
16928 frame rate, only defined for constant frame-rate video
16931 The presentation timestamp in input
16934 The count of the input frame for video or the number of consumed samples,
16935 not including the current frame for audio, starting from 0.
16937 @item NB_CONSUMED_SAMPLES
16938 The number of consumed samples, not including the current frame (only
16941 @item NB_SAMPLES, S
16942 The number of samples in the current frame (only audio)
16944 @item SAMPLE_RATE, SR
16945 The audio sample rate.
16948 The PTS of the first frame.
16951 the time in seconds of the first frame
16954 State whether the current frame is interlaced.
16957 the time in seconds of the current frame
16960 original position in the file of the frame, or undefined if undefined
16961 for the current frame
16964 The previous input PTS.
16967 previous input time in seconds
16970 The previous output PTS.
16973 previous output time in seconds
16976 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
16980 The wallclock (RTC) time at the start of the movie in microseconds.
16983 The timebase of the input timestamps.
16987 @subsection Examples
16991 Start counting PTS from zero
16993 setpts=PTS-STARTPTS
16997 Apply fast motion effect:
17003 Apply slow motion effect:
17009 Set fixed rate of 25 frames per second:
17015 Set fixed rate 25 fps with some jitter:
17017 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17021 Apply an offset of 10 seconds to the input PTS:
17027 Generate timestamps from a "live source" and rebase onto the current timebase:
17029 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17033 Generate timestamps by counting samples:
17040 @section settb, asettb
17042 Set the timebase to use for the output frames timestamps.
17043 It is mainly useful for testing timebase configuration.
17045 It accepts the following parameters:
17050 The expression which is evaluated into the output timebase.
17054 The value for @option{tb} is an arithmetic expression representing a
17055 rational. The expression can contain the constants "AVTB" (the default
17056 timebase), "intb" (the input timebase) and "sr" (the sample rate,
17057 audio only). Default value is "intb".
17059 @subsection Examples
17063 Set the timebase to 1/25:
17069 Set the timebase to 1/10:
17075 Set the timebase to 1001/1000:
17081 Set the timebase to 2*intb:
17087 Set the default timebase value:
17094 Convert input audio to a video output representing frequency spectrum
17095 logarithmically using Brown-Puckette constant Q transform algorithm with
17096 direct frequency domain coefficient calculation (but the transform itself
17097 is not really constant Q, instead the Q factor is actually variable/clamped),
17098 with musical tone scale, from E0 to D#10.
17100 The filter accepts the following options:
17104 Specify the video size for the output. It must be even. For the syntax of this option,
17105 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17106 Default value is @code{1920x1080}.
17109 Set the output frame rate. Default value is @code{25}.
17112 Set the bargraph height. It must be even. Default value is @code{-1} which
17113 computes the bargraph height automatically.
17116 Set the axis height. It must be even. Default value is @code{-1} which computes
17117 the axis height automatically.
17120 Set the sonogram height. It must be even. Default value is @code{-1} which
17121 computes the sonogram height automatically.
17124 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17125 instead. Default value is @code{1}.
17127 @item sono_v, volume
17128 Specify the sonogram volume expression. It can contain variables:
17131 the @var{bar_v} evaluated expression
17132 @item frequency, freq, f
17133 the frequency where it is evaluated
17134 @item timeclamp, tc
17135 the value of @var{timeclamp} option
17139 @item a_weighting(f)
17140 A-weighting of equal loudness
17141 @item b_weighting(f)
17142 B-weighting of equal loudness
17143 @item c_weighting(f)
17144 C-weighting of equal loudness.
17146 Default value is @code{16}.
17148 @item bar_v, volume2
17149 Specify the bargraph volume expression. It can contain variables:
17152 the @var{sono_v} evaluated expression
17153 @item frequency, freq, f
17154 the frequency where it is evaluated
17155 @item timeclamp, tc
17156 the value of @var{timeclamp} option
17160 @item a_weighting(f)
17161 A-weighting of equal loudness
17162 @item b_weighting(f)
17163 B-weighting of equal loudness
17164 @item c_weighting(f)
17165 C-weighting of equal loudness.
17167 Default value is @code{sono_v}.
17169 @item sono_g, gamma
17170 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17171 higher gamma makes the spectrum having more range. Default value is @code{3}.
17172 Acceptable range is @code{[1, 7]}.
17174 @item bar_g, gamma2
17175 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17179 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17180 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17182 @item timeclamp, tc
17183 Specify the transform timeclamp. At low frequency, there is trade-off between
17184 accuracy in time domain and frequency domain. If timeclamp is lower,
17185 event in time domain is represented more accurately (such as fast bass drum),
17186 otherwise event in frequency domain is represented more accurately
17187 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17190 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17191 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17194 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17195 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17198 This option is deprecated and ignored.
17201 Specify the transform length in time domain. Use this option to control accuracy
17202 trade-off between time domain and frequency domain at every frequency sample.
17203 It can contain variables:
17205 @item frequency, freq, f
17206 the frequency where it is evaluated
17207 @item timeclamp, tc
17208 the value of @var{timeclamp} option.
17210 Default value is @code{384*tc/(384+tc*f)}.
17213 Specify the transform count for every video frame. Default value is @code{6}.
17214 Acceptable range is @code{[1, 30]}.
17217 Specify the transform count for every single pixel. Default value is @code{0},
17218 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17221 Specify font file for use with freetype to draw the axis. If not specified,
17222 use embedded font. Note that drawing with font file or embedded font is not
17223 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17227 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17228 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17231 Specify font color expression. This is arithmetic expression that should return
17232 integer value 0xRRGGBB. It can contain variables:
17234 @item frequency, freq, f
17235 the frequency where it is evaluated
17236 @item timeclamp, tc
17237 the value of @var{timeclamp} option
17242 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17243 @item r(x), g(x), b(x)
17244 red, green, and blue value of intensity x.
17246 Default value is @code{st(0, (midi(f)-59.5)/12);
17247 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17248 r(1-ld(1)) + b(ld(1))}.
17251 Specify image file to draw the axis. This option override @var{fontfile} and
17252 @var{fontcolor} option.
17255 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17256 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17257 Default value is @code{1}.
17260 Set colorspace. The accepted values are:
17263 Unspecified (default)
17272 BT.470BG or BT.601-6 625
17275 SMPTE-170M or BT.601-6 525
17281 BT.2020 with non-constant luminance
17286 Set spectrogram color scheme. This is list of floating point values with format
17287 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17288 The default is @code{1|0.5|0|0|0.5|1}.
17292 @subsection Examples
17296 Playing audio while showing the spectrum:
17298 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17302 Same as above, but with frame rate 30 fps:
17304 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17308 Playing at 1280x720:
17310 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17314 Disable sonogram display:
17320 A1 and its harmonics: A1, A2, (near)E3, A3:
17322 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),
17323 asplit[a][out1]; [a] showcqt [out0]'
17327 Same as above, but with more accuracy in frequency domain:
17329 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),
17330 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17336 bar_v=10:sono_v=bar_v*a_weighting(f)
17340 Custom gamma, now spectrum is linear to the amplitude.
17346 Custom tlength equation:
17348 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)))'
17352 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17354 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17358 Custom font using fontconfig:
17360 font='Courier New,Monospace,mono|bold'
17364 Custom frequency range with custom axis using image file:
17366 axisfile=myaxis.png:basefreq=40:endfreq=10000
17372 Convert input audio to video output representing the audio power spectrum.
17373 Audio amplitude is on Y-axis while frequency is on X-axis.
17375 The filter accepts the following options:
17379 Specify size of video. For the syntax of this option, check the
17380 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17381 Default is @code{1024x512}.
17385 This set how each frequency bin will be represented.
17387 It accepts the following values:
17393 Default is @code{bar}.
17396 Set amplitude scale.
17398 It accepts the following values:
17412 Default is @code{log}.
17415 Set frequency scale.
17417 It accepts the following values:
17426 Reverse logarithmic scale.
17428 Default is @code{lin}.
17433 It accepts the following values:
17449 Default is @code{w2048}
17452 Set windowing function.
17454 It accepts the following values:
17476 Default is @code{hanning}.
17479 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17480 which means optimal overlap for selected window function will be picked.
17483 Set time averaging. Setting this to 0 will display current maximal peaks.
17484 Default is @code{1}, which means time averaging is disabled.
17487 Specify list of colors separated by space or by '|' which will be used to
17488 draw channel frequencies. Unrecognized or missing colors will be replaced
17492 Set channel display mode.
17494 It accepts the following values:
17499 Default is @code{combined}.
17502 Set minimum amplitude used in @code{log} amplitude scaler.
17506 @anchor{showspectrum}
17507 @section showspectrum
17509 Convert input audio to a video output, representing the audio frequency
17512 The filter accepts the following options:
17516 Specify the video size for the output. For the syntax of this option, check the
17517 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17518 Default value is @code{640x512}.
17521 Specify how the spectrum should slide along the window.
17523 It accepts the following values:
17526 the samples start again on the left when they reach the right
17528 the samples scroll from right to left
17530 frames are only produced when the samples reach the right
17532 the samples scroll from left to right
17535 Default value is @code{replace}.
17538 Specify display mode.
17540 It accepts the following values:
17543 all channels are displayed in the same row
17545 all channels are displayed in separate rows
17548 Default value is @samp{combined}.
17551 Specify display color mode.
17553 It accepts the following values:
17556 each channel is displayed in a separate color
17558 each channel is displayed using the same color scheme
17560 each channel is displayed using the rainbow color scheme
17562 each channel is displayed using the moreland color scheme
17564 each channel is displayed using the nebulae color scheme
17566 each channel is displayed using the fire color scheme
17568 each channel is displayed using the fiery color scheme
17570 each channel is displayed using the fruit color scheme
17572 each channel is displayed using the cool color scheme
17575 Default value is @samp{channel}.
17578 Specify scale used for calculating intensity color values.
17580 It accepts the following values:
17585 square root, default
17596 Default value is @samp{sqrt}.
17599 Set saturation modifier for displayed colors. Negative values provide
17600 alternative color scheme. @code{0} is no saturation at all.
17601 Saturation must be in [-10.0, 10.0] range.
17602 Default value is @code{1}.
17605 Set window function.
17607 It accepts the following values:
17631 Default value is @code{hann}.
17634 Set orientation of time vs frequency axis. Can be @code{vertical} or
17635 @code{horizontal}. Default is @code{vertical}.
17638 Set ratio of overlap window. Default value is @code{0}.
17639 When value is @code{1} overlap is set to recommended size for specific
17640 window function currently used.
17643 Set scale gain for calculating intensity color values.
17644 Default value is @code{1}.
17647 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17650 Set color rotation, must be in [-1.0, 1.0] range.
17651 Default value is @code{0}.
17654 The usage is very similar to the showwaves filter; see the examples in that
17657 @subsection Examples
17661 Large window with logarithmic color scaling:
17663 showspectrum=s=1280x480:scale=log
17667 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17669 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17670 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17674 @section showspectrumpic
17676 Convert input audio to a single video frame, representing the audio frequency
17679 The filter accepts the following options:
17683 Specify the video size for the output. For the syntax of this option, check the
17684 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17685 Default value is @code{4096x2048}.
17688 Specify display mode.
17690 It accepts the following values:
17693 all channels are displayed in the same row
17695 all channels are displayed in separate rows
17697 Default value is @samp{combined}.
17700 Specify display color mode.
17702 It accepts the following values:
17705 each channel is displayed in a separate color
17707 each channel is displayed using the same color scheme
17709 each channel is displayed using the rainbow color scheme
17711 each channel is displayed using the moreland color scheme
17713 each channel is displayed using the nebulae color scheme
17715 each channel is displayed using the fire color scheme
17717 each channel is displayed using the fiery color scheme
17719 each channel is displayed using the fruit color scheme
17721 each channel is displayed using the cool color scheme
17723 Default value is @samp{intensity}.
17726 Specify scale used for calculating intensity color values.
17728 It accepts the following values:
17733 square root, default
17743 Default value is @samp{log}.
17746 Set saturation modifier for displayed colors. Negative values provide
17747 alternative color scheme. @code{0} is no saturation at all.
17748 Saturation must be in [-10.0, 10.0] range.
17749 Default value is @code{1}.
17752 Set window function.
17754 It accepts the following values:
17777 Default value is @code{hann}.
17780 Set orientation of time vs frequency axis. Can be @code{vertical} or
17781 @code{horizontal}. Default is @code{vertical}.
17784 Set scale gain for calculating intensity color values.
17785 Default value is @code{1}.
17788 Draw time and frequency axes and legends. Default is enabled.
17791 Set color rotation, must be in [-1.0, 1.0] range.
17792 Default value is @code{0}.
17795 @subsection Examples
17799 Extract an audio spectrogram of a whole audio track
17800 in a 1024x1024 picture using @command{ffmpeg}:
17802 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17806 @section showvolume
17808 Convert input audio volume to a video output.
17810 The filter accepts the following options:
17817 Set border width, allowed range is [0, 5]. Default is 1.
17820 Set channel width, allowed range is [80, 8192]. Default is 400.
17823 Set channel height, allowed range is [1, 900]. Default is 20.
17826 Set fade, allowed range is [0.001, 1]. Default is 0.95.
17829 Set volume color expression.
17831 The expression can use the following variables:
17835 Current max volume of channel in dB.
17841 Current channel number, starting from 0.
17845 If set, displays channel names. Default is enabled.
17848 If set, displays volume values. Default is enabled.
17851 Set orientation, can be @code{horizontal} or @code{vertical},
17852 default is @code{horizontal}.
17855 Set step size, allowed range s [0, 5]. Default is 0, which means
17861 Convert input audio to a video output, representing the samples waves.
17863 The filter accepts the following options:
17867 Specify the video size for the output. For the syntax of this option, check the
17868 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17869 Default value is @code{600x240}.
17874 Available values are:
17877 Draw a point for each sample.
17880 Draw a vertical line for each sample.
17883 Draw a point for each sample and a line between them.
17886 Draw a centered vertical line for each sample.
17889 Default value is @code{point}.
17892 Set the number of samples which are printed on the same column. A
17893 larger value will decrease the frame rate. Must be a positive
17894 integer. This option can be set only if the value for @var{rate}
17895 is not explicitly specified.
17898 Set the (approximate) output frame rate. This is done by setting the
17899 option @var{n}. Default value is "25".
17901 @item split_channels
17902 Set if channels should be drawn separately or overlap. Default value is 0.
17905 Set colors separated by '|' which are going to be used for drawing of each channel.
17908 Set amplitude scale.
17910 Available values are:
17928 @subsection Examples
17932 Output the input file audio and the corresponding video representation
17935 amovie=a.mp3,asplit[out0],showwaves[out1]
17939 Create a synthetic signal and show it with showwaves, forcing a
17940 frame rate of 30 frames per second:
17942 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
17946 @section showwavespic
17948 Convert input audio to a single video frame, representing the samples waves.
17950 The filter accepts the following options:
17954 Specify the video size for the output. For the syntax of this option, check the
17955 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17956 Default value is @code{600x240}.
17958 @item split_channels
17959 Set if channels should be drawn separately or overlap. Default value is 0.
17962 Set colors separated by '|' which are going to be used for drawing of each channel.
17965 Set amplitude scale.
17967 Available values are:
17985 @subsection Examples
17989 Extract a channel split representation of the wave form of a whole audio track
17990 in a 1024x800 picture using @command{ffmpeg}:
17992 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
17996 @section sidedata, asidedata
17998 Delete frame side data, or select frames based on it.
18000 This filter accepts the following options:
18004 Set mode of operation of the filter.
18006 Can be one of the following:
18010 Select every frame with side data of @code{type}.
18013 Delete side data of @code{type}. If @code{type} is not set, delete all side
18019 Set side data type used with all modes. Must be set for @code{select} mode. For
18020 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18021 in @file{libavutil/frame.h}. For example, to choose
18022 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18026 @section spectrumsynth
18028 Sythesize audio from 2 input video spectrums, first input stream represents
18029 magnitude across time and second represents phase across time.
18030 The filter will transform from frequency domain as displayed in videos back
18031 to time domain as presented in audio output.
18033 This filter is primarily created for reversing processed @ref{showspectrum}
18034 filter outputs, but can synthesize sound from other spectrograms too.
18035 But in such case results are going to be poor if the phase data is not
18036 available, because in such cases phase data need to be recreated, usually
18037 its just recreated from random noise.
18038 For best results use gray only output (@code{channel} color mode in
18039 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
18040 @code{lin} scale for phase video. To produce phase, for 2nd video, use
18041 @code{data} option. Inputs videos should generally use @code{fullframe}
18042 slide mode as that saves resources needed for decoding video.
18044 The filter accepts the following options:
18048 Specify sample rate of output audio, the sample rate of audio from which
18049 spectrum was generated may differ.
18052 Set number of channels represented in input video spectrums.
18055 Set scale which was used when generating magnitude input spectrum.
18056 Can be @code{lin} or @code{log}. Default is @code{log}.
18059 Set slide which was used when generating inputs spectrums.
18060 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18061 Default is @code{fullframe}.
18064 Set window function used for resynthesis.
18067 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18068 which means optimal overlap for selected window function will be picked.
18071 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18072 Default is @code{vertical}.
18075 @subsection Examples
18079 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18080 then resynthesize videos back to audio with spectrumsynth:
18082 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
18083 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
18084 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18088 @section split, asplit
18090 Split input into several identical outputs.
18092 @code{asplit} works with audio input, @code{split} with video.
18094 The filter accepts a single parameter which specifies the number of outputs. If
18095 unspecified, it defaults to 2.
18097 @subsection Examples
18101 Create two separate outputs from the same input:
18103 [in] split [out0][out1]
18107 To create 3 or more outputs, you need to specify the number of
18110 [in] asplit=3 [out0][out1][out2]
18114 Create two separate outputs from the same input, one cropped and
18117 [in] split [splitout1][splitout2];
18118 [splitout1] crop=100:100:0:0 [cropout];
18119 [splitout2] pad=200:200:100:100 [padout];
18123 Create 5 copies of the input audio with @command{ffmpeg}:
18125 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18131 Receive commands sent through a libzmq client, and forward them to
18132 filters in the filtergraph.
18134 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18135 must be inserted between two video filters, @code{azmq} between two
18138 To enable these filters you need to install the libzmq library and
18139 headers and configure FFmpeg with @code{--enable-libzmq}.
18141 For more information about libzmq see:
18142 @url{http://www.zeromq.org/}
18144 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18145 receives messages sent through a network interface defined by the
18146 @option{bind_address} option.
18148 The received message must be in the form:
18150 @var{TARGET} @var{COMMAND} [@var{ARG}]
18153 @var{TARGET} specifies the target of the command, usually the name of
18154 the filter class or a specific filter instance name.
18156 @var{COMMAND} specifies the name of the command for the target filter.
18158 @var{ARG} is optional and specifies the optional argument list for the
18159 given @var{COMMAND}.
18161 Upon reception, the message is processed and the corresponding command
18162 is injected into the filtergraph. Depending on the result, the filter
18163 will send a reply to the client, adopting the format:
18165 @var{ERROR_CODE} @var{ERROR_REASON}
18169 @var{MESSAGE} is optional.
18171 @subsection Examples
18173 Look at @file{tools/zmqsend} for an example of a zmq client which can
18174 be used to send commands processed by these filters.
18176 Consider the following filtergraph generated by @command{ffplay}
18178 ffplay -dumpgraph 1 -f lavfi "
18179 color=s=100x100:c=red [l];
18180 color=s=100x100:c=blue [r];
18181 nullsrc=s=200x100, zmq [bg];
18182 [bg][l] overlay [bg+l];
18183 [bg+l][r] overlay=x=100 "
18186 To change the color of the left side of the video, the following
18187 command can be used:
18189 echo Parsed_color_0 c yellow | tools/zmqsend
18192 To change the right side:
18194 echo Parsed_color_1 c pink | tools/zmqsend
18197 @c man end MULTIMEDIA FILTERS
18199 @chapter Multimedia Sources
18200 @c man begin MULTIMEDIA SOURCES
18202 Below is a description of the currently available multimedia sources.
18206 This is the same as @ref{movie} source, except it selects an audio
18212 Read audio and/or video stream(s) from a movie container.
18214 It accepts the following parameters:
18218 The name of the resource to read (not necessarily a file; it can also be a
18219 device or a stream accessed through some protocol).
18221 @item format_name, f
18222 Specifies the format assumed for the movie to read, and can be either
18223 the name of a container or an input device. If not specified, the
18224 format is guessed from @var{movie_name} or by probing.
18226 @item seek_point, sp
18227 Specifies the seek point in seconds. The frames will be output
18228 starting from this seek point. The parameter is evaluated with
18229 @code{av_strtod}, so the numerical value may be suffixed by an IS
18230 postfix. The default value is "0".
18233 Specifies the streams to read. Several streams can be specified,
18234 separated by "+". The source will then have as many outputs, in the
18235 same order. The syntax is explained in the ``Stream specifiers''
18236 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18237 respectively the default (best suited) video and audio stream. Default
18238 is "dv", or "da" if the filter is called as "amovie".
18240 @item stream_index, si
18241 Specifies the index of the video stream to read. If the value is -1,
18242 the most suitable video stream will be automatically selected. The default
18243 value is "-1". Deprecated. If the filter is called "amovie", it will select
18244 audio instead of video.
18247 Specifies how many times to read the stream in sequence.
18248 If the value is 0, the stream will be looped infinitely.
18249 Default value is "1".
18251 Note that when the movie is looped the source timestamps are not
18252 changed, so it will generate non monotonically increasing timestamps.
18254 @item discontinuity
18255 Specifies the time difference between frames above which the point is
18256 considered a timestamp discontinuity which is removed by adjusting the later
18260 It allows overlaying a second video on top of the main input of
18261 a filtergraph, as shown in this graph:
18263 input -----------> deltapts0 --> overlay --> output
18266 movie --> scale--> deltapts1 -------+
18268 @subsection Examples
18272 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18273 on top of the input labelled "in":
18275 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18276 [in] setpts=PTS-STARTPTS [main];
18277 [main][over] overlay=16:16 [out]
18281 Read from a video4linux2 device, and overlay it on top of the input
18284 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18285 [in] setpts=PTS-STARTPTS [main];
18286 [main][over] overlay=16:16 [out]
18290 Read the first video stream and the audio stream with id 0x81 from
18291 dvd.vob; the video is connected to the pad named "video" and the audio is
18292 connected to the pad named "audio":
18294 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18298 @subsection Commands
18300 Both movie and amovie support the following commands:
18303 Perform seek using "av_seek_frame".
18304 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18307 @var{stream_index}: If stream_index is -1, a default
18308 stream is selected, and @var{timestamp} is automatically converted
18309 from AV_TIME_BASE units to the stream specific time_base.
18311 @var{timestamp}: Timestamp in AVStream.time_base units
18312 or, if no stream is specified, in AV_TIME_BASE units.
18314 @var{flags}: Flags which select direction and seeking mode.
18318 Get movie duration in AV_TIME_BASE units.
18322 @c man end MULTIMEDIA SOURCES