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 @c man end FILTERGRAPH DESCRIPTION
311 @chapter Audio Filters
312 @c man begin AUDIO FILTERS
314 When you configure your FFmpeg build, you can disable any of the
315 existing filters using @code{--disable-filters}.
316 The configure output will show the audio filters included in your
319 Below is a description of the currently available audio filters.
323 A compressor is mainly used to reduce the dynamic range of a signal.
324 Especially modern music is mostly compressed at a high ratio to
325 improve the overall loudness. It's done to get the highest attention
326 of a listener, "fatten" the sound and bring more "power" to the track.
327 If a signal is compressed too much it may sound dull or "dead"
328 afterwards or it may start to "pump" (which could be a powerful effect
329 but can also destroy a track completely).
330 The right compression is the key to reach a professional sound and is
331 the high art of mixing and mastering. Because of its complex settings
332 it may take a long time to get the right feeling for this kind of effect.
334 Compression is done by detecting the volume above a chosen level
335 @code{threshold} and dividing it by the factor set with @code{ratio}.
336 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
337 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
338 the signal would cause distortion of the waveform the reduction can be
339 levelled over the time. This is done by setting "Attack" and "Release".
340 @code{attack} determines how long the signal has to rise above the threshold
341 before any reduction will occur and @code{release} sets the time the signal
342 has to fall below the threshold to reduce the reduction again. Shorter signals
343 than the chosen attack time will be left untouched.
344 The overall reduction of the signal can be made up afterwards with the
345 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
346 raising the makeup to this level results in a signal twice as loud than the
347 source. To gain a softer entry in the compression the @code{knee} flattens the
348 hard edge at the threshold in the range of the chosen decibels.
350 The filter accepts the following options:
354 If a signal of second stream rises above this level it will affect the gain
355 reduction of the first stream.
356 By default it is 0.125. Range is between 0.00097563 and 1.
359 Set a ratio by which the signal is reduced. 1:2 means that if the level
360 rose 4dB above the threshold, it will be only 2dB above after the reduction.
361 Default is 2. Range is between 1 and 20.
364 Amount of milliseconds the signal has to rise above the threshold before gain
365 reduction starts. Default is 20. Range is between 0.01 and 2000.
368 Amount of milliseconds the signal has to fall below the threshold before
369 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
372 Set the amount by how much signal will be amplified after processing.
373 Default is 2. Range is from 1 and 64.
376 Curve the sharp knee around the threshold to enter gain reduction more softly.
377 Default is 2.82843. Range is between 1 and 8.
380 Choose if the @code{average} level between all channels of input stream
381 or the louder(@code{maximum}) channel of input stream affects the
382 reduction. Default is @code{average}.
385 Should the exact signal be taken in case of @code{peak} or an RMS one in case
386 of @code{rms}. Default is @code{rms} which is mostly smoother.
389 How much to use compressed signal in output. Default is 1.
390 Range is between 0 and 1.
395 Apply cross fade from one input audio stream to another input audio stream.
396 The cross fade is applied for specified duration near the end of first stream.
398 The filter accepts the following options:
402 Specify the number of samples for which the cross fade effect has to last.
403 At the end of the cross fade effect the first input audio will be completely
404 silent. Default is 44100.
407 Specify the duration of the cross fade effect. See
408 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
409 for the accepted syntax.
410 By default the duration is determined by @var{nb_samples}.
411 If set this option is used instead of @var{nb_samples}.
414 Should first stream end overlap with second stream start. Default is enabled.
417 Set curve for cross fade transition for first stream.
420 Set curve for cross fade transition for second stream.
422 For description of available curve types see @ref{afade} filter description.
429 Cross fade from one input to another:
431 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
435 Cross fade from one input to another but without overlapping:
437 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
443 Delay one or more audio channels.
445 Samples in delayed channel are filled with silence.
447 The filter accepts the following option:
451 Set list of delays in milliseconds for each channel separated by '|'.
452 At least one delay greater than 0 should be provided.
453 Unused delays will be silently ignored. If number of given delays is
454 smaller than number of channels all remaining channels will not be delayed.
461 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
462 the second channel (and any other channels that may be present) unchanged.
470 Apply echoing to the input audio.
472 Echoes are reflected sound and can occur naturally amongst mountains
473 (and sometimes large buildings) when talking or shouting; digital echo
474 effects emulate this behaviour and are often used to help fill out the
475 sound of a single instrument or vocal. The time difference between the
476 original signal and the reflection is the @code{delay}, and the
477 loudness of the reflected signal is the @code{decay}.
478 Multiple echoes can have different delays and decays.
480 A description of the accepted parameters follows.
484 Set input gain of reflected signal. Default is @code{0.6}.
487 Set output gain of reflected signal. Default is @code{0.3}.
490 Set list of time intervals in milliseconds between original signal and reflections
491 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
492 Default is @code{1000}.
495 Set list of loudnesses of reflected signals separated by '|'.
496 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
497 Default is @code{0.5}.
504 Make it sound as if there are twice as many instruments as are actually playing:
506 aecho=0.8:0.88:60:0.4
510 If delay is very short, then it sound like a (metallic) robot playing music:
516 A longer delay will sound like an open air concert in the mountains:
518 aecho=0.8:0.9:1000:0.3
522 Same as above but with one more mountain:
524 aecho=0.8:0.9:1000|1800:0.3|0.25
530 Modify an audio signal according to the specified expressions.
532 This filter accepts one or more expressions (one for each channel),
533 which are evaluated and used to modify a corresponding audio signal.
535 It accepts the following parameters:
539 Set the '|'-separated expressions list for each separate channel. If
540 the number of input channels is greater than the number of
541 expressions, the last specified expression is used for the remaining
544 @item channel_layout, c
545 Set output channel layout. If not specified, the channel layout is
546 specified by the number of expressions. If set to @samp{same}, it will
547 use by default the same input channel layout.
550 Each expression in @var{exprs} can contain the following constants and functions:
554 channel number of the current expression
557 number of the evaluated sample, starting from 0
563 time of the evaluated sample expressed in seconds
566 @item nb_out_channels
567 input and output number of channels
570 the value of input channel with number @var{CH}
573 Note: this filter is slow. For faster processing you should use a
582 aeval=val(ch)/2:c=same
586 Invert phase of the second channel:
595 Apply fade-in/out effect to input audio.
597 A description of the accepted parameters follows.
601 Specify the effect type, can be either @code{in} for fade-in, or
602 @code{out} for a fade-out effect. Default is @code{in}.
604 @item start_sample, ss
605 Specify the number of the start sample for starting to apply the fade
606 effect. Default is 0.
609 Specify the number of samples for which the fade effect has to last. At
610 the end of the fade-in effect the output audio will have the same
611 volume as the input audio, at the end of the fade-out transition
612 the output audio will be silence. Default is 44100.
615 Specify the start time of the fade effect. Default is 0.
616 The value must be specified as a time duration; see
617 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
618 for the accepted syntax.
619 If set this option is used instead of @var{start_sample}.
622 Specify the duration of the fade effect. See
623 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
624 for the accepted syntax.
625 At the end of the fade-in effect the output audio will have the same
626 volume as the input audio, at the end of the fade-out transition
627 the output audio will be silence.
628 By default the duration is determined by @var{nb_samples}.
629 If set this option is used instead of @var{nb_samples}.
632 Set curve for fade transition.
634 It accepts the following values:
637 select triangular, linear slope (default)
639 select quarter of sine wave
641 select half of sine wave
643 select exponential sine wave
647 select inverted parabola
661 select inverted quarter of sine wave
663 select inverted half of sine wave
665 select double-exponential seat
667 select double-exponential sigmoid
675 Fade in first 15 seconds of audio:
681 Fade out last 25 seconds of a 900 seconds audio:
683 afade=t=out:st=875:d=25
690 Set output format constraints for the input audio. The framework will
691 negotiate the most appropriate format to minimize conversions.
693 It accepts the following parameters:
697 A '|'-separated list of requested sample formats.
700 A '|'-separated list of requested sample rates.
702 @item channel_layouts
703 A '|'-separated list of requested channel layouts.
705 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
706 for the required syntax.
709 If a parameter is omitted, all values are allowed.
711 Force the output to either unsigned 8-bit or signed 16-bit stereo
713 aformat=sample_fmts=u8|s16:channel_layouts=stereo
718 A gate is mainly used to reduce lower parts of a signal. This kind of signal
719 processing reduces disturbing noise between useful signals.
721 Gating is done by detecting the volume below a chosen level @var{threshold}
722 and divide it by the factor set with @var{ratio}. The bottom of the noise
723 floor is set via @var{range}. Because an exact manipulation of the signal
724 would cause distortion of the waveform the reduction can be levelled over
725 time. This is done by setting @var{attack} and @var{release}.
727 @var{attack} determines how long the signal has to fall below the threshold
728 before any reduction will occur and @var{release} sets the time the signal
729 has to raise above the threshold to reduce the reduction again.
730 Shorter signals than the chosen attack time will be left untouched.
734 Set input level before filtering.
735 Default is 1. Allowed range is from 0.015625 to 64.
738 Set the level of gain reduction when the signal is below the threshold.
739 Default is 0.06125. Allowed range is from 0 to 1.
742 If a signal rises above this level the gain reduction is released.
743 Default is 0.125. Allowed range is from 0 to 1.
746 Set a ratio about which the signal is reduced.
747 Default is 2. Allowed range is from 1 to 9000.
750 Amount of milliseconds the signal has to rise above the threshold before gain
752 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
755 Amount of milliseconds the signal has to fall below the threshold before the
756 reduction is increased again. Default is 250 milliseconds.
757 Allowed range is from 0.01 to 9000.
760 Set amount of amplification of signal after processing.
761 Default is 1. Allowed range is from 1 to 64.
764 Curve the sharp knee around the threshold to enter gain reduction more softly.
765 Default is 2.828427125. Allowed range is from 1 to 8.
768 Choose if exact signal should be taken for detection or an RMS like one.
769 Default is peak. Can be peak or rms.
772 Choose if the average level between all channels or the louder channel affects
774 Default is average. Can be average or maximum.
779 The limiter prevents input signal from raising over a desired threshold.
780 This limiter uses lookahead technology to prevent your signal from distorting.
781 It means that there is a small delay after signal is processed. Keep in mind
782 that the delay it produces is the attack time you set.
784 The filter accepts the following options:
788 Don't let signals above this level pass the limiter. The removed amplitude is
789 added automatically. Default is 1.
792 The limiter will reach its attenuation level in this amount of time in
793 milliseconds. Default is 5 milliseconds.
796 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
797 Default is 50 milliseconds.
800 When gain reduction is always needed ASC takes care of releasing to an
801 average reduction level rather than reaching a reduction of 0 in the release
805 Select how much the release time is affected by ASC, 0 means nearly no changes
806 in release time while 1 produces higher release times.
809 Depending on picked setting it is recommended to upsample input 2x or 4x times
810 with @ref{aresample} before applying this filter.
814 Apply a two-pole all-pass filter with central frequency (in Hz)
815 @var{frequency}, and filter-width @var{width}.
816 An all-pass filter changes the audio's frequency to phase relationship
817 without changing its frequency to amplitude relationship.
819 The filter accepts the following options:
826 Set method to specify band-width of filter.
839 Specify the band-width of a filter in width_type units.
845 Merge two or more audio streams into a single multi-channel stream.
847 The filter accepts the following options:
852 Set the number of inputs. Default is 2.
856 If the channel layouts of the inputs are disjoint, and therefore compatible,
857 the channel layout of the output will be set accordingly and the channels
858 will be reordered as necessary. If the channel layouts of the inputs are not
859 disjoint, the output will have all the channels of the first input then all
860 the channels of the second input, in that order, and the channel layout of
861 the output will be the default value corresponding to the total number of
864 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
865 is FC+BL+BR, then the output will be in 5.1, with the channels in the
866 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
867 first input, b1 is the first channel of the second input).
869 On the other hand, if both input are in stereo, the output channels will be
870 in the default order: a1, a2, b1, b2, and the channel layout will be
871 arbitrarily set to 4.0, which may or may not be the expected value.
873 All inputs must have the same sample rate, and format.
875 If inputs do not have the same duration, the output will stop with the
882 Merge two mono files into a stereo stream:
884 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
888 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
890 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
896 Mixes multiple audio inputs into a single output.
898 Note that this filter only supports float samples (the @var{amerge}
899 and @var{pan} audio filters support many formats). If the @var{amix}
900 input has integer samples then @ref{aresample} will be automatically
901 inserted to perform the conversion to float samples.
905 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
907 will mix 3 input audio streams to a single output with the same duration as the
908 first input and a dropout transition time of 3 seconds.
910 It accepts the following parameters:
914 The number of inputs. If unspecified, it defaults to 2.
917 How to determine the end-of-stream.
921 The duration of the longest input. (default)
924 The duration of the shortest input.
927 The duration of the first input.
931 @item dropout_transition
932 The transition time, in seconds, for volume renormalization when an input
933 stream ends. The default value is 2 seconds.
939 Pass the audio source unchanged to the output.
943 Pad the end of an audio stream with silence.
945 This can be used together with @command{ffmpeg} @option{-shortest} to
946 extend audio streams to the same length as the video stream.
948 A description of the accepted options follows.
952 Set silence packet size. Default value is 4096.
955 Set the number of samples of silence to add to the end. After the
956 value is reached, the stream is terminated. This option is mutually
957 exclusive with @option{whole_len}.
960 Set the minimum total number of samples in the output audio stream. If
961 the value is longer than the input audio length, silence is added to
962 the end, until the value is reached. This option is mutually exclusive
963 with @option{pad_len}.
966 If neither the @option{pad_len} nor the @option{whole_len} option is
967 set, the filter will add silence to the end of the input stream
974 Add 1024 samples of silence to the end of the input:
980 Make sure the audio output will contain at least 10000 samples, pad
981 the input with silence if required:
987 Use @command{ffmpeg} to pad the audio input with silence, so that the
988 video stream will always result the shortest and will be converted
989 until the end in the output file when using the @option{shortest}
992 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
997 Add a phasing effect to the input audio.
999 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1000 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1002 A description of the accepted parameters follows.
1006 Set input gain. Default is 0.4.
1009 Set output gain. Default is 0.74
1012 Set delay in milliseconds. Default is 3.0.
1015 Set decay. Default is 0.4.
1018 Set modulation speed in Hz. Default is 0.5.
1021 Set modulation type. Default is triangular.
1023 It accepts the following values:
1033 Resample the input audio to the specified parameters, using the
1034 libswresample library. If none are specified then the filter will
1035 automatically convert between its input and output.
1037 This filter is also able to stretch/squeeze the audio data to make it match
1038 the timestamps or to inject silence / cut out audio to make it match the
1039 timestamps, do a combination of both or do neither.
1041 The filter accepts the syntax
1042 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1043 expresses a sample rate and @var{resampler_options} is a list of
1044 @var{key}=@var{value} pairs, separated by ":". See the
1045 ffmpeg-resampler manual for the complete list of supported options.
1047 @subsection Examples
1051 Resample the input audio to 44100Hz:
1057 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1058 samples per second compensation:
1060 aresample=async=1000
1064 @section asetnsamples
1066 Set the number of samples per each output audio frame.
1068 The last output packet may contain a different number of samples, as
1069 the filter will flush all the remaining samples when the input audio
1072 The filter accepts the following options:
1076 @item nb_out_samples, n
1077 Set the number of frames per each output audio frame. The number is
1078 intended as the number of samples @emph{per each channel}.
1079 Default value is 1024.
1082 If set to 1, the filter will pad the last audio frame with zeroes, so
1083 that the last frame will contain the same number of samples as the
1084 previous ones. Default value is 1.
1087 For example, to set the number of per-frame samples to 1234 and
1088 disable padding for the last frame, use:
1090 asetnsamples=n=1234:p=0
1095 Set the sample rate without altering the PCM data.
1096 This will result in a change of speed and pitch.
1098 The filter accepts the following options:
1101 @item sample_rate, r
1102 Set the output sample rate. Default is 44100 Hz.
1107 Show a line containing various information for each input audio frame.
1108 The input audio is not modified.
1110 The shown line contains a sequence of key/value pairs of the form
1111 @var{key}:@var{value}.
1113 The following values are shown in the output:
1117 The (sequential) number of the input frame, starting from 0.
1120 The presentation timestamp of the input frame, in time base units; the time base
1121 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1124 The presentation timestamp of the input frame in seconds.
1127 position of the frame in the input stream, -1 if this information in
1128 unavailable and/or meaningless (for example in case of synthetic audio)
1137 The sample rate for the audio frame.
1140 The number of samples (per channel) in the frame.
1143 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1144 audio, the data is treated as if all the planes were concatenated.
1146 @item plane_checksums
1147 A list of Adler-32 checksums for each data plane.
1153 Display time domain statistical information about the audio channels.
1154 Statistics are calculated and displayed for each audio channel and,
1155 where applicable, an overall figure is also given.
1157 It accepts the following option:
1160 Short window length in seconds, used for peak and trough RMS measurement.
1161 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1165 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1166 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1169 Available keys for each channel are:
1200 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1201 this @code{lavfi.astats.Overall.Peak_count}.
1203 For description what each key means read below.
1206 Set number of frame after which stats are going to be recalculated.
1207 Default is disabled.
1210 A description of each shown parameter follows:
1214 Mean amplitude displacement from zero.
1217 Minimal sample level.
1220 Maximal sample level.
1222 @item Min difference
1223 Minimal difference between two consecutive samples.
1225 @item Max difference
1226 Maximal difference between two consecutive samples.
1228 @item Mean difference
1229 Mean difference between two consecutive samples.
1230 The average of each difference between two consecutive samples.
1234 Standard peak and RMS level measured in dBFS.
1238 Peak and trough values for RMS level measured over a short window.
1241 Standard ratio of peak to RMS level (note: not in dB).
1244 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1245 (i.e. either @var{Min level} or @var{Max level}).
1248 Number of occasions (not the number of samples) that the signal attained either
1249 @var{Min level} or @var{Max level}.
1252 Overall bit depth of audio. Number of bits used for each sample.
1257 Synchronize audio data with timestamps by squeezing/stretching it and/or
1258 dropping samples/adding silence when needed.
1260 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1262 It accepts the following parameters:
1266 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1267 by default. When disabled, time gaps are covered with silence.
1270 The minimum difference between timestamps and audio data (in seconds) to trigger
1271 adding/dropping samples. The default value is 0.1. If you get an imperfect
1272 sync with this filter, try setting this parameter to 0.
1275 The maximum compensation in samples per second. Only relevant with compensate=1.
1276 The default value is 500.
1279 Assume that the first PTS should be this value. The time base is 1 / sample
1280 rate. This allows for padding/trimming at the start of the stream. By default,
1281 no assumption is made about the first frame's expected PTS, so no padding or
1282 trimming is done. For example, this could be set to 0 to pad the beginning with
1283 silence if an audio stream starts after the video stream or to trim any samples
1284 with a negative PTS due to encoder delay.
1292 The filter accepts exactly one parameter, the audio tempo. If not
1293 specified then the filter will assume nominal 1.0 tempo. Tempo must
1294 be in the [0.5, 2.0] range.
1296 @subsection Examples
1300 Slow down audio to 80% tempo:
1306 To speed up audio to 125% tempo:
1314 Trim the input so that the output contains one continuous subpart of the input.
1316 It accepts the following parameters:
1319 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1320 sample with the timestamp @var{start} will be the first sample in the output.
1323 Specify time of the first audio sample that will be dropped, i.e. the
1324 audio sample immediately preceding the one with the timestamp @var{end} will be
1325 the last sample in the output.
1328 Same as @var{start}, except this option sets the start timestamp in samples
1332 Same as @var{end}, except this option sets the end timestamp in samples instead
1336 The maximum duration of the output in seconds.
1339 The number of the first sample that should be output.
1342 The number of the first sample that should be dropped.
1345 @option{start}, @option{end}, and @option{duration} are expressed as time
1346 duration specifications; see
1347 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1349 Note that the first two sets of the start/end options and the @option{duration}
1350 option look at the frame timestamp, while the _sample options simply count the
1351 samples that pass through the filter. So start/end_pts and start/end_sample will
1352 give different results when the timestamps are wrong, inexact or do not start at
1353 zero. Also note that this filter does not modify the timestamps. If you wish
1354 to have the output timestamps start at zero, insert the asetpts filter after the
1357 If multiple start or end options are set, this filter tries to be greedy and
1358 keep all samples that match at least one of the specified constraints. To keep
1359 only the part that matches all the constraints at once, chain multiple atrim
1362 The defaults are such that all the input is kept. So it is possible to set e.g.
1363 just the end values to keep everything before the specified time.
1368 Drop everything except the second minute of input:
1370 ffmpeg -i INPUT -af atrim=60:120
1374 Keep only the first 1000 samples:
1376 ffmpeg -i INPUT -af atrim=end_sample=1000
1383 Apply a two-pole Butterworth band-pass filter with central
1384 frequency @var{frequency}, and (3dB-point) band-width width.
1385 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1386 instead of the default: constant 0dB peak gain.
1387 The filter roll off at 6dB per octave (20dB per decade).
1389 The filter accepts the following options:
1393 Set the filter's central frequency. Default is @code{3000}.
1396 Constant skirt gain if set to 1. Defaults to 0.
1399 Set method to specify band-width of filter.
1412 Specify the band-width of a filter in width_type units.
1417 Apply a two-pole Butterworth band-reject filter with central
1418 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1419 The filter roll off at 6dB per octave (20dB per decade).
1421 The filter accepts the following options:
1425 Set the filter's central frequency. Default is @code{3000}.
1428 Set method to specify band-width of filter.
1441 Specify the band-width of a filter in width_type units.
1446 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1447 shelving filter with a response similar to that of a standard
1448 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1450 The filter accepts the following options:
1454 Give the gain at 0 Hz. Its useful range is about -20
1455 (for a large cut) to +20 (for a large boost).
1456 Beware of clipping when using a positive gain.
1459 Set the filter's central frequency and so can be used
1460 to extend or reduce the frequency range to be boosted or cut.
1461 The default value is @code{100} Hz.
1464 Set method to specify band-width of filter.
1477 Determine how steep is the filter's shelf transition.
1482 Apply a biquad IIR filter with the given coefficients.
1483 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1484 are the numerator and denominator coefficients respectively.
1487 Bauer stereo to binaural transformation, which improves headphone listening of
1488 stereo audio records.
1490 It accepts the following parameters:
1494 Pre-defined crossfeed level.
1498 Default level (fcut=700, feed=50).
1501 Chu Moy circuit (fcut=700, feed=60).
1504 Jan Meier circuit (fcut=650, feed=95).
1509 Cut frequency (in Hz).
1518 Remap input channels to new locations.
1520 It accepts the following parameters:
1522 @item channel_layout
1523 The channel layout of the output stream.
1526 Map channels from input to output. The argument is a '|'-separated list of
1527 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1528 @var{in_channel} form. @var{in_channel} can be either the name of the input
1529 channel (e.g. FL for front left) or its index in the input channel layout.
1530 @var{out_channel} is the name of the output channel or its index in the output
1531 channel layout. If @var{out_channel} is not given then it is implicitly an
1532 index, starting with zero and increasing by one for each mapping.
1535 If no mapping is present, the filter will implicitly map input channels to
1536 output channels, preserving indices.
1538 For example, assuming a 5.1+downmix input MOV file,
1540 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1542 will create an output WAV file tagged as stereo from the downmix channels of
1545 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1547 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1550 @section channelsplit
1552 Split each channel from an input audio stream into a separate output stream.
1554 It accepts the following parameters:
1556 @item channel_layout
1557 The channel layout of the input stream. The default is "stereo".
1560 For example, assuming a stereo input MP3 file,
1562 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1564 will create an output Matroska file with two audio streams, one containing only
1565 the left channel and the other the right channel.
1567 Split a 5.1 WAV file into per-channel files:
1569 ffmpeg -i in.wav -filter_complex
1570 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1571 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1572 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1577 Add a chorus effect to the audio.
1579 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1581 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1582 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1583 The modulation depth defines the range the modulated delay is played before or after
1584 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1585 sound tuned around the original one, like in a chorus where some vocals are slightly
1588 It accepts the following parameters:
1591 Set input gain. Default is 0.4.
1594 Set output gain. Default is 0.4.
1597 Set delays. A typical delay is around 40ms to 60ms.
1609 @subsection Examples
1615 chorus=0.7:0.9:55:0.4:0.25:2
1621 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1625 Fuller sounding chorus with three delays:
1627 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
1632 Compress or expand the audio's dynamic range.
1634 It accepts the following parameters:
1640 A list of times in seconds for each channel over which the instantaneous level
1641 of the input signal is averaged to determine its volume. @var{attacks} refers to
1642 increase of volume and @var{decays} refers to decrease of volume. For most
1643 situations, the attack time (response to the audio getting louder) should be
1644 shorter than the decay time, because the human ear is more sensitive to sudden
1645 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1646 a typical value for decay is 0.8 seconds.
1647 If specified number of attacks & decays is lower than number of channels, the last
1648 set attack/decay will be used for all remaining channels.
1651 A list of points for the transfer function, specified in dB relative to the
1652 maximum possible signal amplitude. Each key points list must be defined using
1653 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1654 @code{x0/y0 x1/y1 x2/y2 ....}
1656 The input values must be in strictly increasing order but the transfer function
1657 does not have to be monotonically rising. The point @code{0/0} is assumed but
1658 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1659 function are @code{-70/-70|-60/-20}.
1662 Set the curve radius in dB for all joints. It defaults to 0.01.
1665 Set the additional gain in dB to be applied at all points on the transfer
1666 function. This allows for easy adjustment of the overall gain.
1670 Set an initial volume, in dB, to be assumed for each channel when filtering
1671 starts. This permits the user to supply a nominal level initially, so that, for
1672 example, a very large gain is not applied to initial signal levels before the
1673 companding has begun to operate. A typical value for audio which is initially
1674 quiet is -90 dB. It defaults to 0.
1677 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1678 delayed before being fed to the volume adjuster. Specifying a delay
1679 approximately equal to the attack/decay times allows the filter to effectively
1680 operate in predictive rather than reactive mode. It defaults to 0.
1684 @subsection Examples
1688 Make music with both quiet and loud passages suitable for listening to in a
1691 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1694 Another example for audio with whisper and explosion parts:
1696 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
1700 A noise gate for when the noise is at a lower level than the signal:
1702 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1706 Here is another noise gate, this time for when the noise is at a higher level
1707 than the signal (making it, in some ways, similar to squelch):
1709 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1713 @section compensationdelay
1715 Compensation Delay Line is a metric based delay to compensate differing
1716 positions of microphones or speakers.
1718 For example, you have recorded guitar with two microphones placed in
1719 different location. Because the front of sound wave has fixed speed in
1720 normal conditions, the phasing of microphones can vary and depends on
1721 their location and interposition. The best sound mix can be achieved when
1722 these microphones are in phase (synchronized). Note that distance of
1723 ~30 cm between microphones makes one microphone to capture signal in
1724 antiphase to another microphone. That makes the final mix sounding moody.
1725 This filter helps to solve phasing problems by adding different delays
1726 to each microphone track and make them synchronized.
1728 The best result can be reached when you take one track as base and
1729 synchronize other tracks one by one with it.
1730 Remember that synchronization/delay tolerance depends on sample rate, too.
1731 Higher sample rates will give more tolerance.
1733 It accepts the following parameters:
1737 Set millimeters distance. This is compensation distance for fine tuning.
1741 Set cm distance. This is compensation distance for tightening distance setup.
1745 Set meters distance. This is compensation distance for hard distance setup.
1749 Set dry amount. Amount of unprocessed (dry) signal.
1753 Set wet amount. Amount of processed (wet) signal.
1757 Set temperature degree in Celsius. This is the temperature of the environment.
1762 Apply a DC shift to the audio.
1764 This can be useful to remove a DC offset (caused perhaps by a hardware problem
1765 in the recording chain) from the audio. The effect of a DC offset is reduced
1766 headroom and hence volume. The @ref{astats} filter can be used to determine if
1767 a signal has a DC offset.
1771 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
1775 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
1776 used to prevent clipping.
1780 Dynamic Audio Normalizer.
1782 This filter applies a certain amount of gain to the input audio in order
1783 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
1784 contrast to more "simple" normalization algorithms, the Dynamic Audio
1785 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
1786 This allows for applying extra gain to the "quiet" sections of the audio
1787 while avoiding distortions or clipping the "loud" sections. In other words:
1788 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
1789 sections, in the sense that the volume of each section is brought to the
1790 same target level. Note, however, that the Dynamic Audio Normalizer achieves
1791 this goal *without* applying "dynamic range compressing". It will retain 100%
1792 of the dynamic range *within* each section of the audio file.
1796 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
1797 Default is 500 milliseconds.
1798 The Dynamic Audio Normalizer processes the input audio in small chunks,
1799 referred to as frames. This is required, because a peak magnitude has no
1800 meaning for just a single sample value. Instead, we need to determine the
1801 peak magnitude for a contiguous sequence of sample values. While a "standard"
1802 normalizer would simply use the peak magnitude of the complete file, the
1803 Dynamic Audio Normalizer determines the peak magnitude individually for each
1804 frame. The length of a frame is specified in milliseconds. By default, the
1805 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
1806 been found to give good results with most files.
1807 Note that the exact frame length, in number of samples, will be determined
1808 automatically, based on the sampling rate of the individual input audio file.
1811 Set the Gaussian filter window size. In range from 3 to 301, must be odd
1812 number. Default is 31.
1813 Probably the most important parameter of the Dynamic Audio Normalizer is the
1814 @code{window size} of the Gaussian smoothing filter. The filter's window size
1815 is specified in frames, centered around the current frame. For the sake of
1816 simplicity, this must be an odd number. Consequently, the default value of 31
1817 takes into account the current frame, as well as the 15 preceding frames and
1818 the 15 subsequent frames. Using a larger window results in a stronger
1819 smoothing effect and thus in less gain variation, i.e. slower gain
1820 adaptation. Conversely, using a smaller window results in a weaker smoothing
1821 effect and thus in more gain variation, i.e. faster gain adaptation.
1822 In other words, the more you increase this value, the more the Dynamic Audio
1823 Normalizer will behave like a "traditional" normalization filter. On the
1824 contrary, the more you decrease this value, the more the Dynamic Audio
1825 Normalizer will behave like a dynamic range compressor.
1828 Set the target peak value. This specifies the highest permissible magnitude
1829 level for the normalized audio input. This filter will try to approach the
1830 target peak magnitude as closely as possible, but at the same time it also
1831 makes sure that the normalized signal will never exceed the peak magnitude.
1832 A frame's maximum local gain factor is imposed directly by the target peak
1833 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
1834 It is not recommended to go above this value.
1837 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
1838 The Dynamic Audio Normalizer determines the maximum possible (local) gain
1839 factor for each input frame, i.e. the maximum gain factor that does not
1840 result in clipping or distortion. The maximum gain factor is determined by
1841 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
1842 additionally bounds the frame's maximum gain factor by a predetermined
1843 (global) maximum gain factor. This is done in order to avoid excessive gain
1844 factors in "silent" or almost silent frames. By default, the maximum gain
1845 factor is 10.0, For most inputs the default value should be sufficient and
1846 it usually is not recommended to increase this value. Though, for input
1847 with an extremely low overall volume level, it may be necessary to allow even
1848 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
1849 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
1850 Instead, a "sigmoid" threshold function will be applied. This way, the
1851 gain factors will smoothly approach the threshold value, but never exceed that
1855 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
1856 By default, the Dynamic Audio Normalizer performs "peak" normalization.
1857 This means that the maximum local gain factor for each frame is defined
1858 (only) by the frame's highest magnitude sample. This way, the samples can
1859 be amplified as much as possible without exceeding the maximum signal
1860 level, i.e. without clipping. Optionally, however, the Dynamic Audio
1861 Normalizer can also take into account the frame's root mean square,
1862 abbreviated RMS. In electrical engineering, the RMS is commonly used to
1863 determine the power of a time-varying signal. It is therefore considered
1864 that the RMS is a better approximation of the "perceived loudness" than
1865 just looking at the signal's peak magnitude. Consequently, by adjusting all
1866 frames to a constant RMS value, a uniform "perceived loudness" can be
1867 established. If a target RMS value has been specified, a frame's local gain
1868 factor is defined as the factor that would result in exactly that RMS value.
1869 Note, however, that the maximum local gain factor is still restricted by the
1870 frame's highest magnitude sample, in order to prevent clipping.
1873 Enable channels coupling. By default is enabled.
1874 By default, the Dynamic Audio Normalizer will amplify all channels by the same
1875 amount. This means the same gain factor will be applied to all channels, i.e.
1876 the maximum possible gain factor is determined by the "loudest" channel.
1877 However, in some recordings, it may happen that the volume of the different
1878 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
1879 In this case, this option can be used to disable the channel coupling. This way,
1880 the gain factor will be determined independently for each channel, depending
1881 only on the individual channel's highest magnitude sample. This allows for
1882 harmonizing the volume of the different channels.
1885 Enable DC bias correction. By default is disabled.
1886 An audio signal (in the time domain) is a sequence of sample values.
1887 In the Dynamic Audio Normalizer these sample values are represented in the
1888 -1.0 to 1.0 range, regardless of the original input format. Normally, the
1889 audio signal, or "waveform", should be centered around the zero point.
1890 That means if we calculate the mean value of all samples in a file, or in a
1891 single frame, then the result should be 0.0 or at least very close to that
1892 value. If, however, there is a significant deviation of the mean value from
1893 0.0, in either positive or negative direction, this is referred to as a
1894 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
1895 Audio Normalizer provides optional DC bias correction.
1896 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
1897 the mean value, or "DC correction" offset, of each input frame and subtract
1898 that value from all of the frame's sample values which ensures those samples
1899 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
1900 boundaries, the DC correction offset values will be interpolated smoothly
1901 between neighbouring frames.
1904 Enable alternative boundary mode. By default is disabled.
1905 The Dynamic Audio Normalizer takes into account a certain neighbourhood
1906 around each frame. This includes the preceding frames as well as the
1907 subsequent frames. However, for the "boundary" frames, located at the very
1908 beginning and at the very end of the audio file, not all neighbouring
1909 frames are available. In particular, for the first few frames in the audio
1910 file, the preceding frames are not known. And, similarly, for the last few
1911 frames in the audio file, the subsequent frames are not known. Thus, the
1912 question arises which gain factors should be assumed for the missing frames
1913 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
1914 to deal with this situation. The default boundary mode assumes a gain factor
1915 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
1916 "fade out" at the beginning and at the end of the input, respectively.
1919 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
1920 By default, the Dynamic Audio Normalizer does not apply "traditional"
1921 compression. This means that signal peaks will not be pruned and thus the
1922 full dynamic range will be retained within each local neighbourhood. However,
1923 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
1924 normalization algorithm with a more "traditional" compression.
1925 For this purpose, the Dynamic Audio Normalizer provides an optional compression
1926 (thresholding) function. If (and only if) the compression feature is enabled,
1927 all input frames will be processed by a soft knee thresholding function prior
1928 to the actual normalization process. Put simply, the thresholding function is
1929 going to prune all samples whose magnitude exceeds a certain threshold value.
1930 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
1931 value. Instead, the threshold value will be adjusted for each individual
1933 In general, smaller parameters result in stronger compression, and vice versa.
1934 Values below 3.0 are not recommended, because audible distortion may appear.
1939 Make audio easier to listen to on headphones.
1941 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1942 so that when listened to on headphones the stereo image is moved from
1943 inside your head (standard for headphones) to outside and in front of
1944 the listener (standard for speakers).
1950 Apply a two-pole peaking equalisation (EQ) filter. With this
1951 filter, the signal-level at and around a selected frequency can
1952 be increased or decreased, whilst (unlike bandpass and bandreject
1953 filters) that at all other frequencies is unchanged.
1955 In order to produce complex equalisation curves, this filter can
1956 be given several times, each with a different central frequency.
1958 The filter accepts the following options:
1962 Set the filter's central frequency in Hz.
1965 Set method to specify band-width of filter.
1978 Specify the band-width of a filter in width_type units.
1981 Set the required gain or attenuation in dB.
1982 Beware of clipping when using a positive gain.
1985 @subsection Examples
1988 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1990 equalizer=f=1000:width_type=h:width=200:g=-10
1994 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1996 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2000 @section extrastereo
2002 Linearly increases the difference between left and right channels which
2003 adds some sort of "live" effect to playback.
2005 The filter accepts the following option:
2009 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2010 (average of both channels), with 1.0 sound will be unchanged, with
2011 -1.0 left and right channels will be swapped.
2014 Enable clipping. By default is enabled.
2018 Apply a flanging effect to the audio.
2020 The filter accepts the following options:
2024 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2027 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2030 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2034 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2035 Default value is 71.
2038 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2041 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2042 Default value is @var{sinusoidal}.
2045 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2046 Default value is 25.
2049 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2050 Default is @var{linear}.
2055 Apply a high-pass filter with 3dB point frequency.
2056 The filter can be either single-pole, or double-pole (the default).
2057 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2059 The filter accepts the following options:
2063 Set frequency in Hz. Default is 3000.
2066 Set number of poles. Default is 2.
2069 Set method to specify band-width of filter.
2082 Specify the band-width of a filter in width_type units.
2083 Applies only to double-pole filter.
2084 The default is 0.707q and gives a Butterworth response.
2089 Join multiple input streams into one multi-channel stream.
2091 It accepts the following parameters:
2095 The number of input streams. It defaults to 2.
2097 @item channel_layout
2098 The desired output channel layout. It defaults to stereo.
2101 Map channels from inputs to output. The argument is a '|'-separated list of
2102 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2103 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2104 can be either the name of the input channel (e.g. FL for front left) or its
2105 index in the specified input stream. @var{out_channel} is the name of the output
2109 The filter will attempt to guess the mappings when they are not specified
2110 explicitly. It does so by first trying to find an unused matching input channel
2111 and if that fails it picks the first unused input channel.
2113 Join 3 inputs (with properly set channel layouts):
2115 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2118 Build a 5.1 output from 6 single-channel streams:
2120 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2121 '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'
2127 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2129 To enable compilation of this filter you need to configure FFmpeg with
2130 @code{--enable-ladspa}.
2134 Specifies the name of LADSPA plugin library to load. If the environment
2135 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2136 each one of the directories specified by the colon separated list in
2137 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2138 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2139 @file{/usr/lib/ladspa/}.
2142 Specifies the plugin within the library. Some libraries contain only
2143 one plugin, but others contain many of them. If this is not set filter
2144 will list all available plugins within the specified library.
2147 Set the '|' separated list of controls which are zero or more floating point
2148 values that determine the behavior of the loaded plugin (for example delay,
2150 Controls need to be defined using the following syntax:
2151 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2152 @var{valuei} is the value set on the @var{i}-th control.
2153 Alternatively they can be also defined using the following syntax:
2154 @var{value0}|@var{value1}|@var{value2}|..., where
2155 @var{valuei} is the value set on the @var{i}-th control.
2156 If @option{controls} is set to @code{help}, all available controls and
2157 their valid ranges are printed.
2159 @item sample_rate, s
2160 Specify the sample rate, default to 44100. Only used if plugin have
2164 Set the number of samples per channel per each output frame, default
2165 is 1024. Only used if plugin have zero inputs.
2168 Set the minimum duration of the sourced audio. See
2169 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2170 for the accepted syntax.
2171 Note that the resulting duration may be greater than the specified duration,
2172 as the generated audio is always cut at the end of a complete frame.
2173 If not specified, or the expressed duration is negative, the audio is
2174 supposed to be generated forever.
2175 Only used if plugin have zero inputs.
2179 @subsection Examples
2183 List all available plugins within amp (LADSPA example plugin) library:
2189 List all available controls and their valid ranges for @code{vcf_notch}
2190 plugin from @code{VCF} library:
2192 ladspa=f=vcf:p=vcf_notch:c=help
2196 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2199 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2203 Add reverberation to the audio using TAP-plugins
2204 (Tom's Audio Processing plugins):
2206 ladspa=file=tap_reverb:tap_reverb
2210 Generate white noise, with 0.2 amplitude:
2212 ladspa=file=cmt:noise_source_white:c=c0=.2
2216 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2217 @code{C* Audio Plugin Suite} (CAPS) library:
2219 ladspa=file=caps:Click:c=c1=20'
2223 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2225 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2229 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2230 @code{SWH Plugins} collection:
2232 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2236 Attenuate low frequencies using Multiband EQ from Steve Harris
2237 @code{SWH Plugins} collection:
2239 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2243 @subsection Commands
2245 This filter supports the following commands:
2248 Modify the @var{N}-th control value.
2250 If the specified value is not valid, it is ignored and prior one is kept.
2255 Apply a low-pass filter with 3dB point frequency.
2256 The filter can be either single-pole or double-pole (the default).
2257 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2259 The filter accepts the following options:
2263 Set frequency in Hz. Default is 500.
2266 Set number of poles. Default is 2.
2269 Set method to specify band-width of filter.
2282 Specify the band-width of a filter in width_type units.
2283 Applies only to double-pole filter.
2284 The default is 0.707q and gives a Butterworth response.
2290 Mix channels with specific gain levels. The filter accepts the output
2291 channel layout followed by a set of channels definitions.
2293 This filter is also designed to efficiently remap the channels of an audio
2296 The filter accepts parameters of the form:
2297 "@var{l}|@var{outdef}|@var{outdef}|..."
2301 output channel layout or number of channels
2304 output channel specification, of the form:
2305 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
2308 output channel to define, either a channel name (FL, FR, etc.) or a channel
2309 number (c0, c1, etc.)
2312 multiplicative coefficient for the channel, 1 leaving the volume unchanged
2315 input channel to use, see out_name for details; it is not possible to mix
2316 named and numbered input channels
2319 If the `=' in a channel specification is replaced by `<', then the gains for
2320 that specification will be renormalized so that the total is 1, thus
2321 avoiding clipping noise.
2323 @subsection Mixing examples
2325 For example, if you want to down-mix from stereo to mono, but with a bigger
2326 factor for the left channel:
2328 pan=1c|c0=0.9*c0+0.1*c1
2331 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
2332 7-channels surround:
2334 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
2337 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
2338 that should be preferred (see "-ac" option) unless you have very specific
2341 @subsection Remapping examples
2343 The channel remapping will be effective if, and only if:
2346 @item gain coefficients are zeroes or ones,
2347 @item only one input per channel output,
2350 If all these conditions are satisfied, the filter will notify the user ("Pure
2351 channel mapping detected"), and use an optimized and lossless method to do the
2354 For example, if you have a 5.1 source and want a stereo audio stream by
2355 dropping the extra channels:
2357 pan="stereo| c0=FL | c1=FR"
2360 Given the same source, you can also switch front left and front right channels
2361 and keep the input channel layout:
2363 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
2366 If the input is a stereo audio stream, you can mute the front left channel (and
2367 still keep the stereo channel layout) with:
2372 Still with a stereo audio stream input, you can copy the right channel in both
2373 front left and right:
2375 pan="stereo| c0=FR | c1=FR"
2380 ReplayGain scanner filter. This filter takes an audio stream as an input and
2381 outputs it unchanged.
2382 At end of filtering it displays @code{track_gain} and @code{track_peak}.
2386 Convert the audio sample format, sample rate and channel layout. It is
2387 not meant to be used directly.
2390 Apply time-stretching and pitch-shifting with librubberband.
2392 The filter accepts the following options:
2396 Set tempo scale factor.
2399 Set pitch scale factor.
2402 Set transients detector.
2403 Possible values are:
2412 Possible values are:
2421 Possible values are:
2428 Set processing window size.
2429 Possible values are:
2438 Possible values are:
2445 Enable formant preservation when shift pitching.
2446 Possible values are:
2454 Possible values are:
2463 Possible values are:
2470 @section sidechaincompress
2472 This filter acts like normal compressor but has the ability to compress
2473 detected signal using second input signal.
2474 It needs two input streams and returns one output stream.
2475 First input stream will be processed depending on second stream signal.
2476 The filtered signal then can be filtered with other filters in later stages of
2477 processing. See @ref{pan} and @ref{amerge} filter.
2479 The filter accepts the following options:
2483 If a signal of second stream raises above this level it will affect the gain
2484 reduction of first stream.
2485 By default is 0.125. Range is between 0.00097563 and 1.
2488 Set a ratio about which the signal is reduced. 1:2 means that if the level
2489 raised 4dB above the threshold, it will be only 2dB above after the reduction.
2490 Default is 2. Range is between 1 and 20.
2493 Amount of milliseconds the signal has to rise above the threshold before gain
2494 reduction starts. Default is 20. Range is between 0.01 and 2000.
2497 Amount of milliseconds the signal has to fall below the threshold before
2498 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
2501 Set the amount by how much signal will be amplified after processing.
2502 Default is 2. Range is from 1 and 64.
2505 Curve the sharp knee around the threshold to enter gain reduction more softly.
2506 Default is 2.82843. Range is between 1 and 8.
2509 Choose if the @code{average} level between all channels of side-chain stream
2510 or the louder(@code{maximum}) channel of side-chain stream affects the
2511 reduction. Default is @code{average}.
2514 Should the exact signal be taken in case of @code{peak} or an RMS one in case
2515 of @code{rms}. Default is @code{rms} which is mainly smoother.
2518 How much to use compressed signal in output. Default is 1.
2519 Range is between 0 and 1.
2522 @subsection Examples
2526 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
2527 depending on the signal of 2nd input and later compressed signal to be
2528 merged with 2nd input:
2530 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
2534 @section silencedetect
2536 Detect silence in an audio stream.
2538 This filter logs a message when it detects that the input audio volume is less
2539 or equal to a noise tolerance value for a duration greater or equal to the
2540 minimum detected noise duration.
2542 The printed times and duration are expressed in seconds.
2544 The filter accepts the following options:
2548 Set silence duration until notification (default is 2 seconds).
2551 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
2552 specified value) or amplitude ratio. Default is -60dB, or 0.001.
2555 @subsection Examples
2559 Detect 5 seconds of silence with -50dB noise tolerance:
2561 silencedetect=n=-50dB:d=5
2565 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
2566 tolerance in @file{silence.mp3}:
2568 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
2572 @section silenceremove
2574 Remove silence from the beginning, middle or end of the audio.
2576 The filter accepts the following options:
2580 This value is used to indicate if audio should be trimmed at beginning of
2581 the audio. A value of zero indicates no silence should be trimmed from the
2582 beginning. When specifying a non-zero value, it trims audio up until it
2583 finds non-silence. Normally, when trimming silence from beginning of audio
2584 the @var{start_periods} will be @code{1} but it can be increased to higher
2585 values to trim all audio up to specific count of non-silence periods.
2586 Default value is @code{0}.
2588 @item start_duration
2589 Specify the amount of time that non-silence must be detected before it stops
2590 trimming audio. By increasing the duration, bursts of noises can be treated
2591 as silence and trimmed off. Default value is @code{0}.
2593 @item start_threshold
2594 This indicates what sample value should be treated as silence. For digital
2595 audio, a value of @code{0} may be fine but for audio recorded from analog,
2596 you may wish to increase the value to account for background noise.
2597 Can be specified in dB (in case "dB" is appended to the specified value)
2598 or amplitude ratio. Default value is @code{0}.
2601 Set the count for trimming silence from the end of audio.
2602 To remove silence from the middle of a file, specify a @var{stop_periods}
2603 that is negative. This value is then treated as a positive value and is
2604 used to indicate the effect should restart processing as specified by
2605 @var{start_periods}, making it suitable for removing periods of silence
2606 in the middle of the audio.
2607 Default value is @code{0}.
2610 Specify a duration of silence that must exist before audio is not copied any
2611 more. By specifying a higher duration, silence that is wanted can be left in
2613 Default value is @code{0}.
2615 @item stop_threshold
2616 This is the same as @option{start_threshold} but for trimming silence from
2618 Can be specified in dB (in case "dB" is appended to the specified value)
2619 or amplitude ratio. Default value is @code{0}.
2622 This indicate that @var{stop_duration} length of audio should be left intact
2623 at the beginning of each period of silence.
2624 For example, if you want to remove long pauses between words but do not want
2625 to remove the pauses completely. Default value is @code{0}.
2629 @subsection Examples
2633 The following example shows how this filter can be used to start a recording
2634 that does not contain the delay at the start which usually occurs between
2635 pressing the record button and the start of the performance:
2637 silenceremove=1:5:0.02
2641 @section stereotools
2643 This filter has some handy utilities to manage stereo signals, for converting
2644 M/S stereo recordings to L/R signal while having control over the parameters
2645 or spreading the stereo image of master track.
2647 The filter accepts the following options:
2651 Set input level before filtering for both channels. Defaults is 1.
2652 Allowed range is from 0.015625 to 64.
2655 Set output level after filtering for both channels. Defaults is 1.
2656 Allowed range is from 0.015625 to 64.
2659 Set input balance between both channels. Default is 0.
2660 Allowed range is from -1 to 1.
2663 Set output balance between both channels. Default is 0.
2664 Allowed range is from -1 to 1.
2667 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
2668 clipping. Disabled by default.
2671 Mute the left channel. Disabled by default.
2674 Mute the right channel. Disabled by default.
2677 Change the phase of the left channel. Disabled by default.
2680 Change the phase of the right channel. Disabled by default.
2683 Set stereo mode. Available values are:
2687 Left/Right to Left/Right, this is default.
2690 Left/Right to Mid/Side.
2693 Mid/Side to Left/Right.
2696 Left/Right to Left/Left.
2699 Left/Right to Right/Right.
2702 Left/Right to Left + Right.
2705 Left/Right to Right/Left.
2709 Set level of side signal. Default is 1.
2710 Allowed range is from 0.015625 to 64.
2713 Set balance of side signal. Default is 0.
2714 Allowed range is from -1 to 1.
2717 Set level of the middle signal. Default is 1.
2718 Allowed range is from 0.015625 to 64.
2721 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
2724 Set stereo base between mono and inversed channels. Default is 0.
2725 Allowed range is from -1 to 1.
2728 Set delay in milliseconds how much to delay left from right channel and
2729 vice versa. Default is 0. Allowed range is from -20 to 20.
2732 Set S/C level. Default is 1. Allowed range is from 1 to 100.
2735 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
2738 @section stereowiden
2740 This filter enhance the stereo effect by suppressing signal common to both
2741 channels and by delaying the signal of left into right and vice versa,
2742 thereby widening the stereo effect.
2744 The filter accepts the following options:
2748 Time in milliseconds of the delay of left signal into right and vice versa.
2749 Default is 20 milliseconds.
2752 Amount of gain in delayed signal into right and vice versa. Gives a delay
2753 effect of left signal in right output and vice versa which gives widening
2754 effect. Default is 0.3.
2757 Cross feed of left into right with inverted phase. This helps in suppressing
2758 the mono. If the value is 1 it will cancel all the signal common to both
2759 channels. Default is 0.3.
2762 Set level of input signal of original channel. Default is 0.8.
2767 Boost or cut treble (upper) frequencies of the audio using a two-pole
2768 shelving filter with a response similar to that of a standard
2769 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
2771 The filter accepts the following options:
2775 Give the gain at whichever is the lower of ~22 kHz and the
2776 Nyquist frequency. Its useful range is about -20 (for a large cut)
2777 to +20 (for a large boost). Beware of clipping when using a positive gain.
2780 Set the filter's central frequency and so can be used
2781 to extend or reduce the frequency range to be boosted or cut.
2782 The default value is @code{3000} Hz.
2785 Set method to specify band-width of filter.
2798 Determine how steep is the filter's shelf transition.
2803 Sinusoidal amplitude modulation.
2805 The filter accepts the following options:
2809 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
2810 (20 Hz or lower) will result in a tremolo effect.
2811 This filter may also be used as a ring modulator by specifying
2812 a modulation frequency higher than 20 Hz.
2813 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
2816 Depth of modulation as a percentage. Range is 0.0 - 1.0.
2817 Default value is 0.5.
2822 Sinusoidal phase modulation.
2824 The filter accepts the following options:
2828 Modulation frequency in Hertz.
2829 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
2832 Depth of modulation as a percentage. Range is 0.0 - 1.0.
2833 Default value is 0.5.
2838 Adjust the input audio volume.
2840 It accepts the following parameters:
2844 Set audio volume expression.
2846 Output values are clipped to the maximum value.
2848 The output audio volume is given by the relation:
2850 @var{output_volume} = @var{volume} * @var{input_volume}
2853 The default value for @var{volume} is "1.0".
2856 This parameter represents the mathematical precision.
2858 It determines which input sample formats will be allowed, which affects the
2859 precision of the volume scaling.
2863 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
2865 32-bit floating-point; this limits input sample format to FLT. (default)
2867 64-bit floating-point; this limits input sample format to DBL.
2871 Choose the behaviour on encountering ReplayGain side data in input frames.
2875 Remove ReplayGain side data, ignoring its contents (the default).
2878 Ignore ReplayGain side data, but leave it in the frame.
2881 Prefer the track gain, if present.
2884 Prefer the album gain, if present.
2887 @item replaygain_preamp
2888 Pre-amplification gain in dB to apply to the selected replaygain gain.
2890 Default value for @var{replaygain_preamp} is 0.0.
2893 Set when the volume expression is evaluated.
2895 It accepts the following values:
2898 only evaluate expression once during the filter initialization, or
2899 when the @samp{volume} command is sent
2902 evaluate expression for each incoming frame
2905 Default value is @samp{once}.
2908 The volume expression can contain the following parameters.
2912 frame number (starting at zero)
2915 @item nb_consumed_samples
2916 number of samples consumed by the filter
2918 number of samples in the current frame
2920 original frame position in the file
2926 PTS at start of stream
2928 time at start of stream
2934 last set volume value
2937 Note that when @option{eval} is set to @samp{once} only the
2938 @var{sample_rate} and @var{tb} variables are available, all other
2939 variables will evaluate to NAN.
2941 @subsection Commands
2943 This filter supports the following commands:
2946 Modify the volume expression.
2947 The command accepts the same syntax of the corresponding option.
2949 If the specified expression is not valid, it is kept at its current
2951 @item replaygain_noclip
2952 Prevent clipping by limiting the gain applied.
2954 Default value for @var{replaygain_noclip} is 1.
2958 @subsection Examples
2962 Halve the input audio volume:
2966 volume=volume=-6.0206dB
2969 In all the above example the named key for @option{volume} can be
2970 omitted, for example like in:
2976 Increase input audio power by 6 decibels using fixed-point precision:
2978 volume=volume=6dB:precision=fixed
2982 Fade volume after time 10 with an annihilation period of 5 seconds:
2984 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
2988 @section volumedetect
2990 Detect the volume of the input video.
2992 The filter has no parameters. The input is not modified. Statistics about
2993 the volume will be printed in the log when the input stream end is reached.
2995 In particular it will show the mean volume (root mean square), maximum
2996 volume (on a per-sample basis), and the beginning of a histogram of the
2997 registered volume values (from the maximum value to a cumulated 1/1000 of
3000 All volumes are in decibels relative to the maximum PCM value.
3002 @subsection Examples
3004 Here is an excerpt of the output:
3006 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3007 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3008 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3009 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3010 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3011 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3012 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3013 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3014 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3020 The mean square energy is approximately -27 dB, or 10^-2.7.
3022 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3024 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3027 In other words, raising the volume by +4 dB does not cause any clipping,
3028 raising it by +5 dB causes clipping for 6 samples, etc.
3030 @c man end AUDIO FILTERS
3032 @chapter Audio Sources
3033 @c man begin AUDIO SOURCES
3035 Below is a description of the currently available audio sources.
3039 Buffer audio frames, and make them available to the filter chain.
3041 This source is mainly intended for a programmatic use, in particular
3042 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3044 It accepts the following parameters:
3048 The timebase which will be used for timestamps of submitted frames. It must be
3049 either a floating-point number or in @var{numerator}/@var{denominator} form.
3052 The sample rate of the incoming audio buffers.
3055 The sample format of the incoming audio buffers.
3056 Either a sample format name or its corresponding integer representation from
3057 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3059 @item channel_layout
3060 The channel layout of the incoming audio buffers.
3061 Either a channel layout name from channel_layout_map in
3062 @file{libavutil/channel_layout.c} or its corresponding integer representation
3063 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3066 The number of channels of the incoming audio buffers.
3067 If both @var{channels} and @var{channel_layout} are specified, then they
3072 @subsection Examples
3075 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3078 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3079 Since the sample format with name "s16p" corresponds to the number
3080 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3083 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3088 Generate an audio signal specified by an expression.
3090 This source accepts in input one or more expressions (one for each
3091 channel), which are evaluated and used to generate a corresponding
3094 This source accepts the following options:
3098 Set the '|'-separated expressions list for each separate channel. In case the
3099 @option{channel_layout} option is not specified, the selected channel layout
3100 depends on the number of provided expressions. Otherwise the last
3101 specified expression is applied to the remaining output channels.
3103 @item channel_layout, c
3104 Set the channel layout. The number of channels in the specified layout
3105 must be equal to the number of specified expressions.
3108 Set the minimum duration of the sourced audio. See
3109 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3110 for the accepted syntax.
3111 Note that the resulting duration may be greater than the specified
3112 duration, as the generated audio is always cut at the end of a
3115 If not specified, or the expressed duration is negative, the audio is
3116 supposed to be generated forever.
3119 Set the number of samples per channel per each output frame,
3122 @item sample_rate, s
3123 Specify the sample rate, default to 44100.
3126 Each expression in @var{exprs} can contain the following constants:
3130 number of the evaluated sample, starting from 0
3133 time of the evaluated sample expressed in seconds, starting from 0
3140 @subsection Examples
3150 Generate a sin signal with frequency of 440 Hz, set sample rate to
3153 aevalsrc="sin(440*2*PI*t):s=8000"
3157 Generate a two channels signal, specify the channel layout (Front
3158 Center + Back Center) explicitly:
3160 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
3164 Generate white noise:
3166 aevalsrc="-2+random(0)"
3170 Generate an amplitude modulated signal:
3172 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
3176 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
3178 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
3185 The null audio source, return unprocessed audio frames. It is mainly useful
3186 as a template and to be employed in analysis / debugging tools, or as
3187 the source for filters which ignore the input data (for example the sox
3190 This source accepts the following options:
3194 @item channel_layout, cl
3196 Specifies the channel layout, and can be either an integer or a string
3197 representing a channel layout. The default value of @var{channel_layout}
3200 Check the channel_layout_map definition in
3201 @file{libavutil/channel_layout.c} for the mapping between strings and
3202 channel layout values.
3204 @item sample_rate, r
3205 Specifies the sample rate, and defaults to 44100.
3208 Set the number of samples per requested frames.
3212 @subsection Examples
3216 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
3218 anullsrc=r=48000:cl=4
3222 Do the same operation with a more obvious syntax:
3224 anullsrc=r=48000:cl=mono
3228 All the parameters need to be explicitly defined.
3232 Synthesize a voice utterance using the libflite library.
3234 To enable compilation of this filter you need to configure FFmpeg with
3235 @code{--enable-libflite}.
3237 Note that the flite library is not thread-safe.
3239 The filter accepts the following options:
3244 If set to 1, list the names of the available voices and exit
3245 immediately. Default value is 0.
3248 Set the maximum number of samples per frame. Default value is 512.
3251 Set the filename containing the text to speak.
3254 Set the text to speak.
3257 Set the voice to use for the speech synthesis. Default value is
3258 @code{kal}. See also the @var{list_voices} option.
3261 @subsection Examples
3265 Read from file @file{speech.txt}, and synthesize the text using the
3266 standard flite voice:
3268 flite=textfile=speech.txt
3272 Read the specified text selecting the @code{slt} voice:
3274 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3278 Input text to ffmpeg:
3280 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3284 Make @file{ffplay} speak the specified text, using @code{flite} and
3285 the @code{lavfi} device:
3287 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
3291 For more information about libflite, check:
3292 @url{http://www.speech.cs.cmu.edu/flite/}
3296 Generate a noise audio signal.
3298 The filter accepts the following options:
3301 @item sample_rate, r
3302 Specify the sample rate. Default value is 48000 Hz.
3305 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
3309 Specify the duration of the generated audio stream. Not specifying this option
3310 results in noise with an infinite length.
3312 @item color, colour, c
3313 Specify the color of noise. Available noise colors are white, pink, and brown.
3314 Default color is white.
3317 Specify a value used to seed the PRNG.
3320 Set the number of samples per each output frame, default is 1024.
3323 @subsection Examples
3328 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
3330 anoisesrc=d=60:c=pink:r=44100:a=0.5
3336 Generate an audio signal made of a sine wave with amplitude 1/8.
3338 The audio signal is bit-exact.
3340 The filter accepts the following options:
3345 Set the carrier frequency. Default is 440 Hz.
3347 @item beep_factor, b
3348 Enable a periodic beep every second with frequency @var{beep_factor} times
3349 the carrier frequency. Default is 0, meaning the beep is disabled.
3351 @item sample_rate, r
3352 Specify the sample rate, default is 44100.
3355 Specify the duration of the generated audio stream.
3357 @item samples_per_frame
3358 Set the number of samples per output frame.
3360 The expression can contain the following constants:
3364 The (sequential) number of the output audio frame, starting from 0.
3367 The PTS (Presentation TimeStamp) of the output audio frame,
3368 expressed in @var{TB} units.
3371 The PTS of the output audio frame, expressed in seconds.
3374 The timebase of the output audio frames.
3377 Default is @code{1024}.
3380 @subsection Examples
3385 Generate a simple 440 Hz sine wave:
3391 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
3395 sine=frequency=220:beep_factor=4:duration=5
3399 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
3402 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
3406 @c man end AUDIO SOURCES
3408 @chapter Audio Sinks
3409 @c man begin AUDIO SINKS
3411 Below is a description of the currently available audio sinks.
3413 @section abuffersink
3415 Buffer audio frames, and make them available to the end of filter chain.
3417 This sink is mainly intended for programmatic use, in particular
3418 through the interface defined in @file{libavfilter/buffersink.h}
3419 or the options system.
3421 It accepts a pointer to an AVABufferSinkContext structure, which
3422 defines the incoming buffers' formats, to be passed as the opaque
3423 parameter to @code{avfilter_init_filter} for initialization.
3426 Null audio sink; do absolutely nothing with the input audio. It is
3427 mainly useful as a template and for use in analysis / debugging
3430 @c man end AUDIO SINKS
3432 @chapter Video Filters
3433 @c man begin VIDEO FILTERS
3435 When you configure your FFmpeg build, you can disable any of the
3436 existing filters using @code{--disable-filters}.
3437 The configure output will show the video filters included in your
3440 Below is a description of the currently available video filters.
3442 @section alphaextract
3444 Extract the alpha component from the input as a grayscale video. This
3445 is especially useful with the @var{alphamerge} filter.
3449 Add or replace the alpha component of the primary input with the
3450 grayscale value of a second input. This is intended for use with
3451 @var{alphaextract} to allow the transmission or storage of frame
3452 sequences that have alpha in a format that doesn't support an alpha
3455 For example, to reconstruct full frames from a normal YUV-encoded video
3456 and a separate video created with @var{alphaextract}, you might use:
3458 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
3461 Since this filter is designed for reconstruction, it operates on frame
3462 sequences without considering timestamps, and terminates when either
3463 input reaches end of stream. This will cause problems if your encoding
3464 pipeline drops frames. If you're trying to apply an image as an
3465 overlay to a video stream, consider the @var{overlay} filter instead.
3469 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
3470 and libavformat to work. On the other hand, it is limited to ASS (Advanced
3471 Substation Alpha) subtitles files.
3473 This filter accepts the following option in addition to the common options from
3474 the @ref{subtitles} filter:
3478 Set the shaping engine
3480 Available values are:
3483 The default libass shaping engine, which is the best available.
3485 Fast, font-agnostic shaper that can do only substitutions
3487 Slower shaper using OpenType for substitutions and positioning
3490 The default is @code{auto}.
3494 Apply an Adaptive Temporal Averaging Denoiser to the video input.
3496 The filter accepts the following options:
3500 Set threshold A for 1st plane. Default is 0.02.
3501 Valid range is 0 to 0.3.
3504 Set threshold B for 1st plane. Default is 0.04.
3505 Valid range is 0 to 5.
3508 Set threshold A for 2nd plane. Default is 0.02.
3509 Valid range is 0 to 0.3.
3512 Set threshold B for 2nd plane. Default is 0.04.
3513 Valid range is 0 to 5.
3516 Set threshold A for 3rd plane. Default is 0.02.
3517 Valid range is 0 to 0.3.
3520 Set threshold B for 3rd plane. Default is 0.04.
3521 Valid range is 0 to 5.
3523 Threshold A is designed to react on abrupt changes in the input signal and
3524 threshold B is designed to react on continuous changes in the input signal.
3527 Set number of frames filter will use for averaging. Default is 33. Must be odd
3528 number in range [5, 129].
3533 Compute the bounding box for the non-black pixels in the input frame
3536 This filter computes the bounding box containing all the pixels with a
3537 luminance value greater than the minimum allowed value.
3538 The parameters describing the bounding box are printed on the filter
3541 The filter accepts the following option:
3545 Set the minimal luminance value. Default is @code{16}.
3548 @section blackdetect
3550 Detect video intervals that are (almost) completely black. Can be
3551 useful to detect chapter transitions, commercials, or invalid
3552 recordings. Output lines contains the time for the start, end and
3553 duration of the detected black interval expressed in seconds.
3555 In order to display the output lines, you need to set the loglevel at
3556 least to the AV_LOG_INFO value.
3558 The filter accepts the following options:
3561 @item black_min_duration, d
3562 Set the minimum detected black duration expressed in seconds. It must
3563 be a non-negative floating point number.
3565 Default value is 2.0.
3567 @item picture_black_ratio_th, pic_th
3568 Set the threshold for considering a picture "black".
3569 Express the minimum value for the ratio:
3571 @var{nb_black_pixels} / @var{nb_pixels}
3574 for which a picture is considered black.
3575 Default value is 0.98.
3577 @item pixel_black_th, pix_th
3578 Set the threshold for considering a pixel "black".
3580 The threshold expresses the maximum pixel luminance value for which a
3581 pixel is considered "black". The provided value is scaled according to
3582 the following equation:
3584 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
3587 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
3588 the input video format, the range is [0-255] for YUV full-range
3589 formats and [16-235] for YUV non full-range formats.
3591 Default value is 0.10.
3594 The following example sets the maximum pixel threshold to the minimum
3595 value, and detects only black intervals of 2 or more seconds:
3597 blackdetect=d=2:pix_th=0.00
3602 Detect frames that are (almost) completely black. Can be useful to
3603 detect chapter transitions or commercials. Output lines consist of
3604 the frame number of the detected frame, the percentage of blackness,
3605 the position in the file if known or -1 and the timestamp in seconds.
3607 In order to display the output lines, you need to set the loglevel at
3608 least to the AV_LOG_INFO value.
3610 It accepts the following parameters:
3615 The percentage of the pixels that have to be below the threshold; it defaults to
3618 @item threshold, thresh
3619 The threshold below which a pixel value is considered black; it defaults to
3624 @section blend, tblend
3626 Blend two video frames into each other.
3628 The @code{blend} filter takes two input streams and outputs one
3629 stream, the first input is the "top" layer and second input is
3630 "bottom" layer. Output terminates when shortest input terminates.
3632 The @code{tblend} (time blend) filter takes two consecutive frames
3633 from one single stream, and outputs the result obtained by blending
3634 the new frame on top of the old frame.
3636 A description of the accepted options follows.
3644 Set blend mode for specific pixel component or all pixel components in case
3645 of @var{all_mode}. Default value is @code{normal}.
3647 Available values for component modes are:
3685 Set blend opacity for specific pixel component or all pixel components in case
3686 of @var{all_opacity}. Only used in combination with pixel component blend modes.
3693 Set blend expression for specific pixel component or all pixel components in case
3694 of @var{all_expr}. Note that related mode options will be ignored if those are set.
3696 The expressions can use the following variables:
3700 The sequential number of the filtered frame, starting from @code{0}.
3704 the coordinates of the current sample
3708 the width and height of currently filtered plane
3712 Width and height scale depending on the currently filtered plane. It is the
3713 ratio between the corresponding luma plane number of pixels and the current
3714 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
3715 @code{0.5,0.5} for chroma planes.
3718 Time of the current frame, expressed in seconds.
3721 Value of pixel component at current location for first video frame (top layer).
3724 Value of pixel component at current location for second video frame (bottom layer).
3728 Force termination when the shortest input terminates. Default is
3729 @code{0}. This option is only defined for the @code{blend} filter.
3732 Continue applying the last bottom frame after the end of the stream. A value of
3733 @code{0} disable the filter after the last frame of the bottom layer is reached.
3734 Default is @code{1}. This option is only defined for the @code{blend} filter.
3737 @subsection Examples
3741 Apply transition from bottom layer to top layer in first 10 seconds:
3743 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
3747 Apply 1x1 checkerboard effect:
3749 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
3753 Apply uncover left effect:
3755 blend=all_expr='if(gte(N*SW+X,W),A,B)'
3759 Apply uncover down effect:
3761 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
3765 Apply uncover up-left effect:
3767 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
3771 Display differences between the current and the previous frame:
3773 tblend=all_mode=difference128
3779 Apply a boxblur algorithm to the input video.
3781 It accepts the following parameters:
3785 @item luma_radius, lr
3786 @item luma_power, lp
3787 @item chroma_radius, cr
3788 @item chroma_power, cp
3789 @item alpha_radius, ar
3790 @item alpha_power, ap
3794 A description of the accepted options follows.
3797 @item luma_radius, lr
3798 @item chroma_radius, cr
3799 @item alpha_radius, ar
3800 Set an expression for the box radius in pixels used for blurring the
3801 corresponding input plane.
3803 The radius value must be a non-negative number, and must not be
3804 greater than the value of the expression @code{min(w,h)/2} for the
3805 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
3808 Default value for @option{luma_radius} is "2". If not specified,
3809 @option{chroma_radius} and @option{alpha_radius} default to the
3810 corresponding value set for @option{luma_radius}.
3812 The expressions can contain the following constants:
3816 The input width and height in pixels.
3820 The input chroma image width and height in pixels.
3824 The horizontal and vertical chroma subsample values. For example, for the
3825 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
3828 @item luma_power, lp
3829 @item chroma_power, cp
3830 @item alpha_power, ap
3831 Specify how many times the boxblur filter is applied to the
3832 corresponding plane.
3834 Default value for @option{luma_power} is 2. If not specified,
3835 @option{chroma_power} and @option{alpha_power} default to the
3836 corresponding value set for @option{luma_power}.
3838 A value of 0 will disable the effect.
3841 @subsection Examples
3845 Apply a boxblur filter with the luma, chroma, and alpha radii
3848 boxblur=luma_radius=2:luma_power=1
3853 Set the luma radius to 2, and alpha and chroma radius to 0:
3855 boxblur=2:1:cr=0:ar=0
3859 Set the luma and chroma radii to a fraction of the video dimension:
3861 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
3866 YUV colorspace color/chroma keying.
3868 The filter accepts the following options:
3872 The color which will be replaced with transparency.
3875 Similarity percentage with the key color.
3877 0.01 matches only the exact key color, while 1.0 matches everything.
3882 0.0 makes pixels either fully transparent, or not transparent at all.
3884 Higher values result in semi-transparent pixels, with a higher transparency
3885 the more similar the pixels color is to the key color.
3888 Signals that the color passed is already in YUV instead of RGB.
3890 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
3891 This can be used to pass exact YUV values as hexadecimal numbers.
3894 @subsection Examples
3898 Make every green pixel in the input image transparent:
3900 ffmpeg -i input.png -vf chromakey=green out.png
3904 Overlay a greenscreen-video on top of a static black background.
3906 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
3912 Visualize information exported by some codecs.
3914 Some codecs can export information through frames using side-data or other
3915 means. For example, some MPEG based codecs export motion vectors through the
3916 @var{export_mvs} flag in the codec @option{flags2} option.
3918 The filter accepts the following option:
3922 Set motion vectors to visualize.
3924 Available flags for @var{mv} are:
3928 forward predicted MVs of P-frames
3930 forward predicted MVs of B-frames
3932 backward predicted MVs of B-frames
3936 @subsection Examples
3940 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
3942 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
3946 @section colorbalance
3947 Modify intensity of primary colors (red, green and blue) of input frames.
3949 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
3950 regions for the red-cyan, green-magenta or blue-yellow balance.
3952 A positive adjustment value shifts the balance towards the primary color, a negative
3953 value towards the complementary color.
3955 The filter accepts the following options:
3961 Adjust red, green and blue shadows (darkest pixels).
3966 Adjust red, green and blue midtones (medium pixels).
3971 Adjust red, green and blue highlights (brightest pixels).
3973 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3976 @subsection Examples
3980 Add red color cast to shadows:
3987 RGB colorspace color keying.
3989 The filter accepts the following options:
3993 The color which will be replaced with transparency.
3996 Similarity percentage with the key color.
3998 0.01 matches only the exact key color, while 1.0 matches everything.
4003 0.0 makes pixels either fully transparent, or not transparent at all.
4005 Higher values result in semi-transparent pixels, with a higher transparency
4006 the more similar the pixels color is to the key color.
4009 @subsection Examples
4013 Make every green pixel in the input image transparent:
4015 ffmpeg -i input.png -vf colorkey=green out.png
4019 Overlay a greenscreen-video on top of a static background image.
4021 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
4025 @section colorlevels
4027 Adjust video input frames using levels.
4029 The filter accepts the following options:
4036 Adjust red, green, blue and alpha input black point.
4037 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4043 Adjust red, green, blue and alpha input white point.
4044 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
4046 Input levels are used to lighten highlights (bright tones), darken shadows
4047 (dark tones), change the balance of bright and dark tones.
4053 Adjust red, green, blue and alpha output black point.
4054 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
4060 Adjust red, green, blue and alpha output white point.
4061 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
4063 Output levels allows manual selection of a constrained output level range.
4066 @subsection Examples
4070 Make video output darker:
4072 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
4078 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
4082 Make video output lighter:
4084 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
4088 Increase brightness:
4090 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
4094 @section colorchannelmixer
4096 Adjust video input frames by re-mixing color channels.
4098 This filter modifies a color channel by adding the values associated to
4099 the other channels of the same pixels. For example if the value to
4100 modify is red, the output value will be:
4102 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
4105 The filter accepts the following options:
4112 Adjust contribution of input red, green, blue and alpha channels for output red channel.
4113 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
4119 Adjust contribution of input red, green, blue and alpha channels for output green channel.
4120 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
4126 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
4127 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
4133 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
4134 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
4136 Allowed ranges for options are @code{[-2.0, 2.0]}.
4139 @subsection Examples
4143 Convert source to grayscale:
4145 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
4148 Simulate sepia tones:
4150 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
4154 @section colormatrix
4156 Convert color matrix.
4158 The filter accepts the following options:
4163 Specify the source and destination color matrix. Both values must be
4166 The accepted values are:
4182 For example to convert from BT.601 to SMPTE-240M, use the command:
4184 colormatrix=bt601:smpte240m
4189 Copy the input source unchanged to the output. This is mainly useful for
4194 Crop the input video to given dimensions.
4196 It accepts the following parameters:
4200 The width of the output video. It defaults to @code{iw}.
4201 This expression is evaluated only once during the filter
4202 configuration, or when the @samp{w} or @samp{out_w} command is sent.
4205 The height of the output video. It defaults to @code{ih}.
4206 This expression is evaluated only once during the filter
4207 configuration, or when the @samp{h} or @samp{out_h} command is sent.
4210 The horizontal position, in the input video, of the left edge of the output
4211 video. It defaults to @code{(in_w-out_w)/2}.
4212 This expression is evaluated per-frame.
4215 The vertical position, in the input video, of the top edge of the output video.
4216 It defaults to @code{(in_h-out_h)/2}.
4217 This expression is evaluated per-frame.
4220 If set to 1 will force the output display aspect ratio
4221 to be the same of the input, by changing the output sample aspect
4222 ratio. It defaults to 0.
4225 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
4226 expressions containing the following constants:
4231 The computed values for @var{x} and @var{y}. They are evaluated for
4236 The input width and height.
4240 These are the same as @var{in_w} and @var{in_h}.
4244 The output (cropped) width and height.
4248 These are the same as @var{out_w} and @var{out_h}.
4251 same as @var{iw} / @var{ih}
4254 input sample aspect ratio
4257 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
4261 horizontal and vertical chroma subsample values. For example for the
4262 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4265 The number of the input frame, starting from 0.
4268 the position in the file of the input frame, NAN if unknown
4271 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
4275 The expression for @var{out_w} may depend on the value of @var{out_h},
4276 and the expression for @var{out_h} may depend on @var{out_w}, but they
4277 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
4278 evaluated after @var{out_w} and @var{out_h}.
4280 The @var{x} and @var{y} parameters specify the expressions for the
4281 position of the top-left corner of the output (non-cropped) area. They
4282 are evaluated for each frame. If the evaluated value is not valid, it
4283 is approximated to the nearest valid value.
4285 The expression for @var{x} may depend on @var{y}, and the expression
4286 for @var{y} may depend on @var{x}.
4288 @subsection Examples
4292 Crop area with size 100x100 at position (12,34).
4297 Using named options, the example above becomes:
4299 crop=w=100:h=100:x=12:y=34
4303 Crop the central input area with size 100x100:
4309 Crop the central input area with size 2/3 of the input video:
4311 crop=2/3*in_w:2/3*in_h
4315 Crop the input video central square:
4322 Delimit the rectangle with the top-left corner placed at position
4323 100:100 and the right-bottom corner corresponding to the right-bottom
4324 corner of the input image.
4326 crop=in_w-100:in_h-100:100:100
4330 Crop 10 pixels from the left and right borders, and 20 pixels from
4331 the top and bottom borders
4333 crop=in_w-2*10:in_h-2*20
4337 Keep only the bottom right quarter of the input image:
4339 crop=in_w/2:in_h/2:in_w/2:in_h/2
4343 Crop height for getting Greek harmony:
4345 crop=in_w:1/PHI*in_w
4349 Apply trembling effect:
4351 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)
4355 Apply erratic camera effect depending on timestamp:
4357 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)"
4361 Set x depending on the value of y:
4363 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
4367 @subsection Commands
4369 This filter supports the following commands:
4375 Set width/height of the output video and the horizontal/vertical position
4377 The command accepts the same syntax of the corresponding option.
4379 If the specified expression is not valid, it is kept at its current
4385 Auto-detect the crop size.
4387 It calculates the necessary cropping parameters and prints the
4388 recommended parameters via the logging system. The detected dimensions
4389 correspond to the non-black area of the input video.
4391 It accepts the following parameters:
4396 Set higher black value threshold, which can be optionally specified
4397 from nothing (0) to everything (255 for 8bit based formats). An intensity
4398 value greater to the set value is considered non-black. It defaults to 24.
4399 You can also specify a value between 0.0 and 1.0 which will be scaled depending
4400 on the bitdepth of the pixel format.
4403 The value which the width/height should be divisible by. It defaults to
4404 16. The offset is automatically adjusted to center the video. Use 2 to
4405 get only even dimensions (needed for 4:2:2 video). 16 is best when
4406 encoding to most video codecs.
4408 @item reset_count, reset
4409 Set the counter that determines after how many frames cropdetect will
4410 reset the previously detected largest video area and start over to
4411 detect the current optimal crop area. Default value is 0.
4413 This can be useful when channel logos distort the video area. 0
4414 indicates 'never reset', and returns the largest area encountered during
4421 Apply color adjustments using curves.
4423 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
4424 component (red, green and blue) has its values defined by @var{N} key points
4425 tied from each other using a smooth curve. The x-axis represents the pixel
4426 values from the input frame, and the y-axis the new pixel values to be set for
4429 By default, a component curve is defined by the two points @var{(0;0)} and
4430 @var{(1;1)}. This creates a straight line where each original pixel value is
4431 "adjusted" to its own value, which means no change to the image.
4433 The filter allows you to redefine these two points and add some more. A new
4434 curve (using a natural cubic spline interpolation) will be define to pass
4435 smoothly through all these new coordinates. The new defined points needs to be
4436 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
4437 be in the @var{[0;1]} interval. If the computed curves happened to go outside
4438 the vector spaces, the values will be clipped accordingly.
4440 If there is no key point defined in @code{x=0}, the filter will automatically
4441 insert a @var{(0;0)} point. In the same way, if there is no key point defined
4442 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
4444 The filter accepts the following options:
4448 Select one of the available color presets. This option can be used in addition
4449 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
4450 options takes priority on the preset values.
4451 Available presets are:
4454 @item color_negative
4457 @item increase_contrast
4459 @item linear_contrast
4460 @item medium_contrast
4462 @item strong_contrast
4465 Default is @code{none}.
4467 Set the master key points. These points will define a second pass mapping. It
4468 is sometimes called a "luminance" or "value" mapping. It can be used with
4469 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
4470 post-processing LUT.
4472 Set the key points for the red component.
4474 Set the key points for the green component.
4476 Set the key points for the blue component.
4478 Set the key points for all components (not including master).
4479 Can be used in addition to the other key points component
4480 options. In this case, the unset component(s) will fallback on this
4481 @option{all} setting.
4483 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
4486 To avoid some filtergraph syntax conflicts, each key points list need to be
4487 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
4489 @subsection Examples
4493 Increase slightly the middle level of blue:
4495 curves=blue='0.5/0.58'
4501 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
4503 Here we obtain the following coordinates for each components:
4506 @code{(0;0.11) (0.42;0.51) (1;0.95)}
4508 @code{(0;0) (0.50;0.48) (1;1)}
4510 @code{(0;0.22) (0.49;0.44) (1;0.80)}
4514 The previous example can also be achieved with the associated built-in preset:
4516 curves=preset=vintage
4526 Use a Photoshop preset and redefine the points of the green component:
4528 curves=psfile='MyCurvesPresets/purple.acv':green='0.45/0.53'
4534 Denoise frames using 2D DCT (frequency domain filtering).
4536 This filter is not designed for real time.
4538 The filter accepts the following options:
4542 Set the noise sigma constant.
4544 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
4545 coefficient (absolute value) below this threshold with be dropped.
4547 If you need a more advanced filtering, see @option{expr}.
4549 Default is @code{0}.
4552 Set number overlapping pixels for each block. Since the filter can be slow, you
4553 may want to reduce this value, at the cost of a less effective filter and the
4554 risk of various artefacts.
4556 If the overlapping value doesn't permit processing the whole input width or
4557 height, a warning will be displayed and according borders won't be denoised.
4559 Default value is @var{blocksize}-1, which is the best possible setting.
4562 Set the coefficient factor expression.
4564 For each coefficient of a DCT block, this expression will be evaluated as a
4565 multiplier value for the coefficient.
4567 If this is option is set, the @option{sigma} option will be ignored.
4569 The absolute value of the coefficient can be accessed through the @var{c}
4573 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
4574 @var{blocksize}, which is the width and height of the processed blocks.
4576 The default value is @var{3} (8x8) and can be raised to @var{4} for a
4577 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
4578 on the speed processing. Also, a larger block size does not necessarily means a
4582 @subsection Examples
4584 Apply a denoise with a @option{sigma} of @code{4.5}:
4589 The same operation can be achieved using the expression system:
4591 dctdnoiz=e='gte(c, 4.5*3)'
4594 Violent denoise using a block size of @code{16x16}:
4601 Remove banding artifacts from input video.
4602 It works by replacing banded pixels with average value of referenced pixels.
4604 The filter accepts the following options:
4611 Set banding detection threshold for each plane. Default is 0.02.
4612 Valid range is 0.00003 to 0.5.
4613 If difference between current pixel and reference pixel is less than threshold,
4614 it will be considered as banded.
4617 Banding detection range in pixels. Default is 16. If positive, random number
4618 in range 0 to set value will be used. If negative, exact absolute value
4620 The range defines square of four pixels around current pixel.
4623 Set direction in radians from which four pixel will be compared. If positive,
4624 random direction from 0 to set direction will be picked. If negative, exact of
4625 absolute value will be picked. For example direction 0, -PI or -2*PI radians
4626 will pick only pixels on same row and -PI/2 will pick only pixels on same
4630 If enabled, current pixel is compared with average value of all four
4631 surrounding pixels. The default is enabled. If disabled current pixel is
4632 compared with all four surrounding pixels. The pixel is considered banded
4633 if only all four differences with surrounding pixels are less than threshold.
4639 Drop duplicated frames at regular intervals.
4641 The filter accepts the following options:
4645 Set the number of frames from which one will be dropped. Setting this to
4646 @var{N} means one frame in every batch of @var{N} frames will be dropped.
4647 Default is @code{5}.
4650 Set the threshold for duplicate detection. If the difference metric for a frame
4651 is less than or equal to this value, then it is declared as duplicate. Default
4655 Set scene change threshold. Default is @code{15}.
4659 Set the size of the x and y-axis blocks used during metric calculations.
4660 Larger blocks give better noise suppression, but also give worse detection of
4661 small movements. Must be a power of two. Default is @code{32}.
4664 Mark main input as a pre-processed input and activate clean source input
4665 stream. This allows the input to be pre-processed with various filters to help
4666 the metrics calculation while keeping the frame selection lossless. When set to
4667 @code{1}, the first stream is for the pre-processed input, and the second
4668 stream is the clean source from where the kept frames are chosen. Default is
4672 Set whether or not chroma is considered in the metric calculations. Default is
4678 Apply deflate effect to the video.
4680 This filter replaces the pixel by the local(3x3) average by taking into account
4681 only values lower than the pixel.
4683 It accepts the following options:
4690 Limit the maximum change for each plane, default is 65535.
4691 If 0, plane will remain unchanged.
4696 Remove judder produced by partially interlaced telecined content.
4698 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
4699 source was partially telecined content then the output of @code{pullup,dejudder}
4700 will have a variable frame rate. May change the recorded frame rate of the
4701 container. Aside from that change, this filter will not affect constant frame
4704 The option available in this filter is:
4708 Specify the length of the window over which the judder repeats.
4710 Accepts any integer greater than 1. Useful values are:
4714 If the original was telecined from 24 to 30 fps (Film to NTSC).
4717 If the original was telecined from 25 to 30 fps (PAL to NTSC).
4720 If a mixture of the two.
4723 The default is @samp{4}.
4728 Suppress a TV station logo by a simple interpolation of the surrounding
4729 pixels. Just set a rectangle covering the logo and watch it disappear
4730 (and sometimes something even uglier appear - your mileage may vary).
4732 It accepts the following parameters:
4737 Specify the top left corner coordinates of the logo. They must be
4742 Specify the width and height of the logo to clear. They must be
4746 Specify the thickness of the fuzzy edge of the rectangle (added to
4747 @var{w} and @var{h}). The default value is 1. This option is
4748 deprecated, setting higher values should no longer be necessary and
4752 When set to 1, a green rectangle is drawn on the screen to simplify
4753 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
4754 The default value is 0.
4756 The rectangle is drawn on the outermost pixels which will be (partly)
4757 replaced with interpolated values. The values of the next pixels
4758 immediately outside this rectangle in each direction will be used to
4759 compute the interpolated pixel values inside the rectangle.
4763 @subsection Examples
4767 Set a rectangle covering the area with top left corner coordinates 0,0
4768 and size 100x77, and a band of size 10:
4770 delogo=x=0:y=0:w=100:h=77:band=10
4777 Attempt to fix small changes in horizontal and/or vertical shift. This
4778 filter helps remove camera shake from hand-holding a camera, bumping a
4779 tripod, moving on a vehicle, etc.
4781 The filter accepts the following options:
4789 Specify a rectangular area where to limit the search for motion
4791 If desired the search for motion vectors can be limited to a
4792 rectangular area of the frame defined by its top left corner, width
4793 and height. These parameters have the same meaning as the drawbox
4794 filter which can be used to visualise the position of the bounding
4797 This is useful when simultaneous movement of subjects within the frame
4798 might be confused for camera motion by the motion vector search.
4800 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
4801 then the full frame is used. This allows later options to be set
4802 without specifying the bounding box for the motion vector search.
4804 Default - search the whole frame.
4808 Specify the maximum extent of movement in x and y directions in the
4809 range 0-64 pixels. Default 16.
4812 Specify how to generate pixels to fill blanks at the edge of the
4813 frame. Available values are:
4816 Fill zeroes at blank locations
4818 Original image at blank locations
4820 Extruded edge value at blank locations
4822 Mirrored edge at blank locations
4824 Default value is @samp{mirror}.
4827 Specify the blocksize to use for motion search. Range 4-128 pixels,
4831 Specify the contrast threshold for blocks. Only blocks with more than
4832 the specified contrast (difference between darkest and lightest
4833 pixels) will be considered. Range 1-255, default 125.
4836 Specify the search strategy. Available values are:
4839 Set exhaustive search
4841 Set less exhaustive search.
4843 Default value is @samp{exhaustive}.
4846 If set then a detailed log of the motion search is written to the
4850 If set to 1, specify using OpenCL capabilities, only available if
4851 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
4857 Apply an exact inverse of the telecine operation. It requires a predefined
4858 pattern specified using the pattern option which must be the same as that passed
4859 to the telecine filter.
4861 This filter accepts the following options:
4870 The default value is @code{top}.
4874 A string of numbers representing the pulldown pattern you wish to apply.
4875 The default value is @code{23}.
4878 A number representing position of the first frame with respect to the telecine
4879 pattern. This is to be used if the stream is cut. The default value is @code{0}.
4884 Apply dilation effect to the video.
4886 This filter replaces the pixel by the local(3x3) maximum.
4888 It accepts the following options:
4895 Limit the maximum change for each plane, default is 65535.
4896 If 0, plane will remain unchanged.
4899 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
4902 Flags to local 3x3 coordinates maps like this:
4911 Displace pixels as indicated by second and third input stream.
4913 It takes three input streams and outputs one stream, the first input is the
4914 source, and second and third input are displacement maps.
4916 The second input specifies how much to displace pixels along the
4917 x-axis, while the third input specifies how much to displace pixels
4919 If one of displacement map streams terminates, last frame from that
4920 displacement map will be used.
4922 Note that once generated, displacements maps can be reused over and over again.
4924 A description of the accepted options follows.
4928 Set displace behavior for pixels that are out of range.
4930 Available values are:
4933 Missing pixels are replaced by black pixels.
4936 Adjacent pixels will spread out to replace missing pixels.
4939 Out of range pixels are wrapped so they point to pixels of other side.
4941 Default is @samp{smear}.
4945 @subsection Examples
4949 Add ripple effect to rgb input of video size hd720:
4951 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
4955 Add wave effect to rgb input of video size hd720:
4957 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
4963 Draw a colored box on the input image.
4965 It accepts the following parameters:
4970 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
4974 The expressions which specify the width and height of the box; if 0 they are interpreted as
4975 the input width and height. It defaults to 0.
4978 Specify the color of the box to write. For the general syntax of this option,
4979 check the "Color" section in the ffmpeg-utils manual. If the special
4980 value @code{invert} is used, the box edge color is the same as the
4981 video with inverted luma.
4984 The expression which sets the thickness of the box edge. Default value is @code{3}.
4986 See below for the list of accepted constants.
4989 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
4990 following constants:
4994 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
4998 horizontal and vertical chroma subsample values. For example for the
4999 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5003 The input width and height.
5006 The input sample aspect ratio.
5010 The x and y offset coordinates where the box is drawn.
5014 The width and height of the drawn box.
5017 The thickness of the drawn box.
5019 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
5020 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
5024 @subsection Examples
5028 Draw a black box around the edge of the input image:
5034 Draw a box with color red and an opacity of 50%:
5036 drawbox=10:20:200:60:red@@0.5
5039 The previous example can be specified as:
5041 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
5045 Fill the box with pink color:
5047 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
5051 Draw a 2-pixel red 2.40:1 mask:
5053 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
5057 @section drawgraph, adrawgraph
5059 Draw a graph using input video or audio metadata.
5061 It accepts the following parameters:
5065 Set 1st frame metadata key from which metadata values will be used to draw a graph.
5068 Set 1st foreground color expression.
5071 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
5074 Set 2nd foreground color expression.
5077 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
5080 Set 3rd foreground color expression.
5083 Set 4th frame metadata key from which metadata values will be used to draw a graph.
5086 Set 4th foreground color expression.
5089 Set minimal value of metadata value.
5092 Set maximal value of metadata value.
5095 Set graph background color. Default is white.
5100 Available values for mode is:
5107 Default is @code{line}.
5112 Available values for slide is:
5115 Draw new frame when right border is reached.
5118 Replace old columns with new ones.
5121 Scroll from right to left.
5124 Scroll from left to right.
5127 Default is @code{frame}.
5130 Set size of graph video. For the syntax of this option, check the
5131 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
5132 The default value is @code{900x256}.
5134 The foreground color expressions can use the following variables:
5137 Minimal value of metadata value.
5140 Maximal value of metadata value.
5143 Current metadata key value.
5146 The color is defined as 0xAABBGGRR.
5149 Example using metadata from @ref{signalstats} filter:
5151 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
5154 Example using metadata from @ref{ebur128} filter:
5156 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
5161 Draw a grid on the input image.
5163 It accepts the following parameters:
5168 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
5172 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
5173 input width and height, respectively, minus @code{thickness}, so image gets
5174 framed. Default to 0.
5177 Specify the color of the grid. For the general syntax of this option,
5178 check the "Color" section in the ffmpeg-utils manual. If the special
5179 value @code{invert} is used, the grid color is the same as the
5180 video with inverted luma.
5183 The expression which sets the thickness of the grid line. Default value is @code{1}.
5185 See below for the list of accepted constants.
5188 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
5189 following constants:
5193 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
5197 horizontal and vertical chroma subsample values. For example for the
5198 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5202 The input grid cell width and height.
5205 The input sample aspect ratio.
5209 The x and y coordinates of some point of grid intersection (meant to configure offset).
5213 The width and height of the drawn cell.
5216 The thickness of the drawn cell.
5218 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
5219 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
5223 @subsection Examples
5227 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
5229 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
5233 Draw a white 3x3 grid with an opacity of 50%:
5235 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
5242 Draw a text string or text from a specified file on top of a video, using the
5243 libfreetype library.
5245 To enable compilation of this filter, you need to configure FFmpeg with
5246 @code{--enable-libfreetype}.
5247 To enable default font fallback and the @var{font} option you need to
5248 configure FFmpeg with @code{--enable-libfontconfig}.
5249 To enable the @var{text_shaping} option, you need to configure FFmpeg with
5250 @code{--enable-libfribidi}.
5254 It accepts the following parameters:
5259 Used to draw a box around text using the background color.
5260 The value must be either 1 (enable) or 0 (disable).
5261 The default value of @var{box} is 0.
5264 Set the width of the border to be drawn around the box using @var{boxcolor}.
5265 The default value of @var{boxborderw} is 0.
5268 The color to be used for drawing box around text. For the syntax of this
5269 option, check the "Color" section in the ffmpeg-utils manual.
5271 The default value of @var{boxcolor} is "white".
5274 Set the width of the border to be drawn around the text using @var{bordercolor}.
5275 The default value of @var{borderw} is 0.
5278 Set the color to be used for drawing border around text. For the syntax of this
5279 option, check the "Color" section in the ffmpeg-utils manual.
5281 The default value of @var{bordercolor} is "black".
5284 Select how the @var{text} is expanded. Can be either @code{none},
5285 @code{strftime} (deprecated) or
5286 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
5290 If true, check and fix text coords to avoid clipping.
5293 The color to be used for drawing fonts. For the syntax of this option, check
5294 the "Color" section in the ffmpeg-utils manual.
5296 The default value of @var{fontcolor} is "black".
5298 @item fontcolor_expr
5299 String which is expanded the same way as @var{text} to obtain dynamic
5300 @var{fontcolor} value. By default this option has empty value and is not
5301 processed. When this option is set, it overrides @var{fontcolor} option.
5304 The font family to be used for drawing text. By default Sans.
5307 The font file to be used for drawing text. The path must be included.
5308 This parameter is mandatory if the fontconfig support is disabled.
5311 This option does not exist, please see the timeline system
5314 Draw the text applying alpha blending. The value can
5315 be either a number between 0.0 and 1.0
5316 The expression accepts the same variables @var{x, y} do.
5317 The default value is 1.
5318 Please see fontcolor_expr
5321 The font size to be used for drawing text.
5322 The default value of @var{fontsize} is 16.
5325 If set to 1, attempt to shape the text (for example, reverse the order of
5326 right-to-left text and join Arabic characters) before drawing it.
5327 Otherwise, just draw the text exactly as given.
5328 By default 1 (if supported).
5331 The flags to be used for loading the fonts.
5333 The flags map the corresponding flags supported by libfreetype, and are
5334 a combination of the following values:
5341 @item vertical_layout
5342 @item force_autohint
5345 @item ignore_global_advance_width
5347 @item ignore_transform
5353 Default value is "default".
5355 For more information consult the documentation for the FT_LOAD_*
5359 The color to be used for drawing a shadow behind the drawn text. For the
5360 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
5362 The default value of @var{shadowcolor} is "black".
5366 The x and y offsets for the text shadow position with respect to the
5367 position of the text. They can be either positive or negative
5368 values. The default value for both is "0".
5371 The starting frame number for the n/frame_num variable. The default value
5375 The size in number of spaces to use for rendering the tab.
5379 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
5380 format. It can be used with or without text parameter. @var{timecode_rate}
5381 option must be specified.
5383 @item timecode_rate, rate, r
5384 Set the timecode frame rate (timecode only).
5387 The text string to be drawn. The text must be a sequence of UTF-8
5389 This parameter is mandatory if no file is specified with the parameter
5393 A text file containing text to be drawn. The text must be a sequence
5394 of UTF-8 encoded characters.
5396 This parameter is mandatory if no text string is specified with the
5397 parameter @var{text}.
5399 If both @var{text} and @var{textfile} are specified, an error is thrown.
5402 If set to 1, the @var{textfile} will be reloaded before each frame.
5403 Be sure to update it atomically, or it may be read partially, or even fail.
5407 The expressions which specify the offsets where text will be drawn
5408 within the video frame. They are relative to the top/left border of the
5411 The default value of @var{x} and @var{y} is "0".
5413 See below for the list of accepted constants and functions.
5416 The parameters for @var{x} and @var{y} are expressions containing the
5417 following constants and functions:
5421 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
5425 horizontal and vertical chroma subsample values. For example for the
5426 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5429 the height of each text line
5437 @item max_glyph_a, ascent
5438 the maximum distance from the baseline to the highest/upper grid
5439 coordinate used to place a glyph outline point, for all the rendered
5441 It is a positive value, due to the grid's orientation with the Y axis
5444 @item max_glyph_d, descent
5445 the maximum distance from the baseline to the lowest grid coordinate
5446 used to place a glyph outline point, for all the rendered glyphs.
5447 This is a negative value, due to the grid's orientation, with the Y axis
5451 maximum glyph height, that is the maximum height for all the glyphs
5452 contained in the rendered text, it is equivalent to @var{ascent} -
5456 maximum glyph width, that is the maximum width for all the glyphs
5457 contained in the rendered text
5460 the number of input frame, starting from 0
5462 @item rand(min, max)
5463 return a random number included between @var{min} and @var{max}
5466 The input sample aspect ratio.
5469 timestamp expressed in seconds, NAN if the input timestamp is unknown
5472 the height of the rendered text
5475 the width of the rendered text
5479 the x and y offset coordinates where the text is drawn.
5481 These parameters allow the @var{x} and @var{y} expressions to refer
5482 each other, so you can for example specify @code{y=x/dar}.
5485 @anchor{drawtext_expansion}
5486 @subsection Text expansion
5488 If @option{expansion} is set to @code{strftime},
5489 the filter recognizes strftime() sequences in the provided text and
5490 expands them accordingly. Check the documentation of strftime(). This
5491 feature is deprecated.
5493 If @option{expansion} is set to @code{none}, the text is printed verbatim.
5495 If @option{expansion} is set to @code{normal} (which is the default),
5496 the following expansion mechanism is used.
5498 The backslash character @samp{\}, followed by any character, always expands to
5499 the second character.
5501 Sequence of the form @code{%@{...@}} are expanded. The text between the
5502 braces is a function name, possibly followed by arguments separated by ':'.
5503 If the arguments contain special characters or delimiters (':' or '@}'),
5504 they should be escaped.
5506 Note that they probably must also be escaped as the value for the
5507 @option{text} option in the filter argument string and as the filter
5508 argument in the filtergraph description, and possibly also for the shell,
5509 that makes up to four levels of escaping; using a text file avoids these
5512 The following functions are available:
5517 The expression evaluation result.
5519 It must take one argument specifying the expression to be evaluated,
5520 which accepts the same constants and functions as the @var{x} and
5521 @var{y} values. Note that not all constants should be used, for
5522 example the text size is not known when evaluating the expression, so
5523 the constants @var{text_w} and @var{text_h} will have an undefined
5526 @item expr_int_format, eif
5527 Evaluate the expression's value and output as formatted integer.
5529 The first argument is the expression to be evaluated, just as for the @var{expr} function.
5530 The second argument specifies the output format. Allowed values are @samp{x},
5531 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
5532 @code{printf} function.
5533 The third parameter is optional and sets the number of positions taken by the output.
5534 It can be used to add padding with zeros from the left.
5537 The time at which the filter is running, expressed in UTC.
5538 It can accept an argument: a strftime() format string.
5541 The time at which the filter is running, expressed in the local time zone.
5542 It can accept an argument: a strftime() format string.
5545 Frame metadata. It must take one argument specifying metadata key.
5548 The frame number, starting from 0.
5551 A 1 character description of the current picture type.
5554 The timestamp of the current frame.
5555 It can take up to three arguments.
5557 The first argument is the format of the timestamp; it defaults to @code{flt}
5558 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
5559 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
5560 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
5561 @code{localtime} stands for the timestamp of the frame formatted as
5562 local time zone time.
5564 The second argument is an offset added to the timestamp.
5566 If the format is set to @code{localtime} or @code{gmtime},
5567 a third argument may be supplied: a strftime() format string.
5568 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
5571 @subsection Examples
5575 Draw "Test Text" with font FreeSerif, using the default values for the
5576 optional parameters.
5579 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
5583 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
5584 and y=50 (counting from the top-left corner of the screen), text is
5585 yellow with a red box around it. Both the text and the box have an
5589 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
5590 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
5593 Note that the double quotes are not necessary if spaces are not used
5594 within the parameter list.
5597 Show the text at the center of the video frame:
5599 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
5603 Show a text line sliding from right to left in the last row of the video
5604 frame. The file @file{LONG_LINE} is assumed to contain a single line
5607 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
5611 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
5613 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
5617 Draw a single green letter "g", at the center of the input video.
5618 The glyph baseline is placed at half screen height.
5620 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
5624 Show text for 1 second every 3 seconds:
5626 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
5630 Use fontconfig to set the font. Note that the colons need to be escaped.
5632 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
5636 Print the date of a real-time encoding (see strftime(3)):
5638 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
5642 Show text fading in and out (appearing/disappearing):
5645 DS=1.0 # display start
5646 DE=10.0 # display end
5647 FID=1.5 # fade in duration
5648 FOD=5 # fade out duration
5649 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 @}"
5654 For more information about libfreetype, check:
5655 @url{http://www.freetype.org/}.
5657 For more information about fontconfig, check:
5658 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
5660 For more information about libfribidi, check:
5661 @url{http://fribidi.org/}.
5665 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
5667 The filter accepts the following options:
5672 Set low and high threshold values used by the Canny thresholding
5675 The high threshold selects the "strong" edge pixels, which are then
5676 connected through 8-connectivity with the "weak" edge pixels selected
5677 by the low threshold.
5679 @var{low} and @var{high} threshold values must be chosen in the range
5680 [0,1], and @var{low} should be lesser or equal to @var{high}.
5682 Default value for @var{low} is @code{20/255}, and default value for @var{high}
5686 Define the drawing mode.
5690 Draw white/gray wires on black background.
5693 Mix the colors to create a paint/cartoon effect.
5696 Default value is @var{wires}.
5699 @subsection Examples
5703 Standard edge detection with custom values for the hysteresis thresholding:
5705 edgedetect=low=0.1:high=0.4
5709 Painting effect without thresholding:
5711 edgedetect=mode=colormix:high=0
5716 Set brightness, contrast, saturation and approximate gamma adjustment.
5718 The filter accepts the following options:
5722 Set the contrast expression. The value must be a float value in range
5723 @code{-2.0} to @code{2.0}. The default value is "1".
5726 Set the brightness expression. The value must be a float value in
5727 range @code{-1.0} to @code{1.0}. The default value is "0".
5730 Set the saturation expression. The value must be a float in
5731 range @code{0.0} to @code{3.0}. The default value is "1".
5734 Set the gamma expression. The value must be a float in range
5735 @code{0.1} to @code{10.0}. The default value is "1".
5738 Set the gamma expression for red. The value must be a float in
5739 range @code{0.1} to @code{10.0}. The default value is "1".
5742 Set the gamma expression for green. The value must be a float in range
5743 @code{0.1} to @code{10.0}. The default value is "1".
5746 Set the gamma expression for blue. The value must be a float in range
5747 @code{0.1} to @code{10.0}. The default value is "1".
5750 Set the gamma weight expression. It can be used to reduce the effect
5751 of a high gamma value on bright image areas, e.g. keep them from
5752 getting overamplified and just plain white. The value must be a float
5753 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
5754 gamma correction all the way down while @code{1.0} leaves it at its
5755 full strength. Default is "1".
5758 Set when the expressions for brightness, contrast, saturation and
5759 gamma expressions are evaluated.
5761 It accepts the following values:
5764 only evaluate expressions once during the filter initialization or
5765 when a command is processed
5768 evaluate expressions for each incoming frame
5771 Default value is @samp{init}.
5774 The expressions accept the following parameters:
5777 frame count of the input frame starting from 0
5780 byte position of the corresponding packet in the input file, NAN if
5784 frame rate of the input video, NAN if the input frame rate is unknown
5787 timestamp expressed in seconds, NAN if the input timestamp is unknown
5790 @subsection Commands
5791 The filter supports the following commands:
5795 Set the contrast expression.
5798 Set the brightness expression.
5801 Set the saturation expression.
5804 Set the gamma expression.
5807 Set the gamma_r expression.
5810 Set gamma_g expression.
5813 Set gamma_b expression.
5816 Set gamma_weight expression.
5818 The command accepts the same syntax of the corresponding option.
5820 If the specified expression is not valid, it is kept at its current
5827 Apply erosion effect to the video.
5829 This filter replaces the pixel by the local(3x3) minimum.
5831 It accepts the following options:
5838 Limit the maximum change for each plane, default is 65535.
5839 If 0, plane will remain unchanged.
5842 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
5845 Flags to local 3x3 coordinates maps like this:
5852 @section extractplanes
5854 Extract color channel components from input video stream into
5855 separate grayscale video streams.
5857 The filter accepts the following option:
5861 Set plane(s) to extract.
5863 Available values for planes are:
5874 Choosing planes not available in the input will result in an error.
5875 That means you cannot select @code{r}, @code{g}, @code{b} planes
5876 with @code{y}, @code{u}, @code{v} planes at same time.
5879 @subsection Examples
5883 Extract luma, u and v color channel component from input video frame
5884 into 3 grayscale outputs:
5886 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
5892 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
5894 For each input image, the filter will compute the optimal mapping from
5895 the input to the output given the codebook length, that is the number
5896 of distinct output colors.
5898 This filter accepts the following options.
5901 @item codebook_length, l
5902 Set codebook length. The value must be a positive integer, and
5903 represents the number of distinct output colors. Default value is 256.
5906 Set the maximum number of iterations to apply for computing the optimal
5907 mapping. The higher the value the better the result and the higher the
5908 computation time. Default value is 1.
5911 Set a random seed, must be an integer included between 0 and
5912 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
5913 will try to use a good random seed on a best effort basis.
5916 Set pal8 output pixel format. This option does not work with codebook
5917 length greater than 256.
5922 Apply a fade-in/out effect to the input video.
5924 It accepts the following parameters:
5928 The effect type can be either "in" for a fade-in, or "out" for a fade-out
5930 Default is @code{in}.
5932 @item start_frame, s
5933 Specify the number of the frame to start applying the fade
5934 effect at. Default is 0.
5937 The number of frames that the fade effect lasts. At the end of the
5938 fade-in effect, the output video will have the same intensity as the input video.
5939 At the end of the fade-out transition, the output video will be filled with the
5940 selected @option{color}.
5944 If set to 1, fade only alpha channel, if one exists on the input.
5947 @item start_time, st
5948 Specify the timestamp (in seconds) of the frame to start to apply the fade
5949 effect. If both start_frame and start_time are specified, the fade will start at
5950 whichever comes last. Default is 0.
5953 The number of seconds for which the fade effect has to last. At the end of the
5954 fade-in effect the output video will have the same intensity as the input video,
5955 at the end of the fade-out transition the output video will be filled with the
5956 selected @option{color}.
5957 If both duration and nb_frames are specified, duration is used. Default is 0
5958 (nb_frames is used by default).
5961 Specify the color of the fade. Default is "black".
5964 @subsection Examples
5968 Fade in the first 30 frames of video:
5973 The command above is equivalent to:
5979 Fade out the last 45 frames of a 200-frame video:
5982 fade=type=out:start_frame=155:nb_frames=45
5986 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
5988 fade=in:0:25, fade=out:975:25
5992 Make the first 5 frames yellow, then fade in from frame 5-24:
5994 fade=in:5:20:color=yellow
5998 Fade in alpha over first 25 frames of video:
6000 fade=in:0:25:alpha=1
6004 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
6006 fade=t=in:st=5.5:d=0.5
6012 Apply arbitrary expressions to samples in frequency domain
6016 Adjust the dc value (gain) of the luma plane of the image. The filter
6017 accepts an integer value in range @code{0} to @code{1000}. The default
6018 value is set to @code{0}.
6021 Adjust the dc value (gain) of the 1st chroma plane of the image. The
6022 filter accepts an integer value in range @code{0} to @code{1000}. The
6023 default value is set to @code{0}.
6026 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
6027 filter accepts an integer value in range @code{0} to @code{1000}. The
6028 default value is set to @code{0}.
6031 Set the frequency domain weight expression for the luma plane.
6034 Set the frequency domain weight expression for the 1st chroma plane.
6037 Set the frequency domain weight expression for the 2nd chroma plane.
6039 The filter accepts the following variables:
6042 The coordinates of the current sample.
6046 The width and height of the image.
6049 @subsection Examples
6055 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
6061 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
6067 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
6074 Extract a single field from an interlaced image using stride
6075 arithmetic to avoid wasting CPU time. The output frames are marked as
6078 The filter accepts the following options:
6082 Specify whether to extract the top (if the value is @code{0} or
6083 @code{top}) or the bottom field (if the value is @code{1} or
6089 Field matching filter for inverse telecine. It is meant to reconstruct the
6090 progressive frames from a telecined stream. The filter does not drop duplicated
6091 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
6092 followed by a decimation filter such as @ref{decimate} in the filtergraph.
6094 The separation of the field matching and the decimation is notably motivated by
6095 the possibility of inserting a de-interlacing filter fallback between the two.
6096 If the source has mixed telecined and real interlaced content,
6097 @code{fieldmatch} will not be able to match fields for the interlaced parts.
6098 But these remaining combed frames will be marked as interlaced, and thus can be
6099 de-interlaced by a later filter such as @ref{yadif} before decimation.
6101 In addition to the various configuration options, @code{fieldmatch} can take an
6102 optional second stream, activated through the @option{ppsrc} option. If
6103 enabled, the frames reconstruction will be based on the fields and frames from
6104 this second stream. This allows the first input to be pre-processed in order to
6105 help the various algorithms of the filter, while keeping the output lossless
6106 (assuming the fields are matched properly). Typically, a field-aware denoiser,
6107 or brightness/contrast adjustments can help.
6109 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
6110 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
6111 which @code{fieldmatch} is based on. While the semantic and usage are very
6112 close, some behaviour and options names can differ.
6114 The @ref{decimate} filter currently only works for constant frame rate input.
6115 If your input has mixed telecined (30fps) and progressive content with a lower
6116 framerate like 24fps use the following filterchain to produce the necessary cfr
6117 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
6119 The filter accepts the following options:
6123 Specify the assumed field order of the input stream. Available values are:
6127 Auto detect parity (use FFmpeg's internal parity value).
6129 Assume bottom field first.
6131 Assume top field first.
6134 Note that it is sometimes recommended not to trust the parity announced by the
6137 Default value is @var{auto}.
6140 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
6141 sense that it won't risk creating jerkiness due to duplicate frames when
6142 possible, but if there are bad edits or blended fields it will end up
6143 outputting combed frames when a good match might actually exist. On the other
6144 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
6145 but will almost always find a good frame if there is one. The other values are
6146 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
6147 jerkiness and creating duplicate frames versus finding good matches in sections
6148 with bad edits, orphaned fields, blended fields, etc.
6150 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
6152 Available values are:
6156 2-way matching (p/c)
6158 2-way matching, and trying 3rd match if still combed (p/c + n)
6160 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
6162 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
6163 still combed (p/c + n + u/b)
6165 3-way matching (p/c/n)
6167 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
6168 detected as combed (p/c/n + u/b)
6171 The parenthesis at the end indicate the matches that would be used for that
6172 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
6175 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
6178 Default value is @var{pc_n}.
6181 Mark the main input stream as a pre-processed input, and enable the secondary
6182 input stream as the clean source to pick the fields from. See the filter
6183 introduction for more details. It is similar to the @option{clip2} feature from
6186 Default value is @code{0} (disabled).
6189 Set the field to match from. It is recommended to set this to the same value as
6190 @option{order} unless you experience matching failures with that setting. In
6191 certain circumstances changing the field that is used to match from can have a
6192 large impact on matching performance. Available values are:
6196 Automatic (same value as @option{order}).
6198 Match from the bottom field.
6200 Match from the top field.
6203 Default value is @var{auto}.
6206 Set whether or not chroma is included during the match comparisons. In most
6207 cases it is recommended to leave this enabled. You should set this to @code{0}
6208 only if your clip has bad chroma problems such as heavy rainbowing or other
6209 artifacts. Setting this to @code{0} could also be used to speed things up at
6210 the cost of some accuracy.
6212 Default value is @code{1}.
6216 These define an exclusion band which excludes the lines between @option{y0} and
6217 @option{y1} from being included in the field matching decision. An exclusion
6218 band can be used to ignore subtitles, a logo, or other things that may
6219 interfere with the matching. @option{y0} sets the starting scan line and
6220 @option{y1} sets the ending line; all lines in between @option{y0} and
6221 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
6222 @option{y0} and @option{y1} to the same value will disable the feature.
6223 @option{y0} and @option{y1} defaults to @code{0}.
6226 Set the scene change detection threshold as a percentage of maximum change on
6227 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
6228 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
6229 @option{scthresh} is @code{[0.0, 100.0]}.
6231 Default value is @code{12.0}.
6234 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
6235 account the combed scores of matches when deciding what match to use as the
6236 final match. Available values are:
6240 No final matching based on combed scores.
6242 Combed scores are only used when a scene change is detected.
6244 Use combed scores all the time.
6247 Default is @var{sc}.
6250 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
6251 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
6252 Available values are:
6256 No forced calculation.
6258 Force p/c/n calculations.
6260 Force p/c/n/u/b calculations.
6263 Default value is @var{none}.
6266 This is the area combing threshold used for combed frame detection. This
6267 essentially controls how "strong" or "visible" combing must be to be detected.
6268 Larger values mean combing must be more visible and smaller values mean combing
6269 can be less visible or strong and still be detected. Valid settings are from
6270 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
6271 be detected as combed). This is basically a pixel difference value. A good
6272 range is @code{[8, 12]}.
6274 Default value is @code{9}.
6277 Sets whether or not chroma is considered in the combed frame decision. Only
6278 disable this if your source has chroma problems (rainbowing, etc.) that are
6279 causing problems for the combed frame detection with chroma enabled. Actually,
6280 using @option{chroma}=@var{0} is usually more reliable, except for the case
6281 where there is chroma only combing in the source.
6283 Default value is @code{0}.
6287 Respectively set the x-axis and y-axis size of the window used during combed
6288 frame detection. This has to do with the size of the area in which
6289 @option{combpel} pixels are required to be detected as combed for a frame to be
6290 declared combed. See the @option{combpel} parameter description for more info.
6291 Possible values are any number that is a power of 2 starting at 4 and going up
6294 Default value is @code{16}.
6297 The number of combed pixels inside any of the @option{blocky} by
6298 @option{blockx} size blocks on the frame for the frame to be detected as
6299 combed. While @option{cthresh} controls how "visible" the combing must be, this
6300 setting controls "how much" combing there must be in any localized area (a
6301 window defined by the @option{blockx} and @option{blocky} settings) on the
6302 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
6303 which point no frames will ever be detected as combed). This setting is known
6304 as @option{MI} in TFM/VFM vocabulary.
6306 Default value is @code{80}.
6309 @anchor{p/c/n/u/b meaning}
6310 @subsection p/c/n/u/b meaning
6312 @subsubsection p/c/n
6314 We assume the following telecined stream:
6317 Top fields: 1 2 2 3 4
6318 Bottom fields: 1 2 3 4 4
6321 The numbers correspond to the progressive frame the fields relate to. Here, the
6322 first two frames are progressive, the 3rd and 4th are combed, and so on.
6324 When @code{fieldmatch} is configured to run a matching from bottom
6325 (@option{field}=@var{bottom}) this is how this input stream get transformed:
6330 B 1 2 3 4 4 <-- matching reference
6339 As a result of the field matching, we can see that some frames get duplicated.
6340 To perform a complete inverse telecine, you need to rely on a decimation filter
6341 after this operation. See for instance the @ref{decimate} filter.
6343 The same operation now matching from top fields (@option{field}=@var{top})
6348 T 1 2 2 3 4 <-- matching reference
6358 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
6359 basically, they refer to the frame and field of the opposite parity:
6362 @item @var{p} matches the field of the opposite parity in the previous frame
6363 @item @var{c} matches the field of the opposite parity in the current frame
6364 @item @var{n} matches the field of the opposite parity in the next frame
6369 The @var{u} and @var{b} matching are a bit special in the sense that they match
6370 from the opposite parity flag. In the following examples, we assume that we are
6371 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
6372 'x' is placed above and below each matched fields.
6374 With bottom matching (@option{field}=@var{bottom}):
6379 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6380 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6388 With top matching (@option{field}=@var{top}):
6393 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6394 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6402 @subsection Examples
6404 Simple IVTC of a top field first telecined stream:
6406 fieldmatch=order=tff:combmatch=none, decimate
6409 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
6411 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
6416 Transform the field order of the input video.
6418 It accepts the following parameters:
6423 The output field order. Valid values are @var{tff} for top field first or @var{bff}
6424 for bottom field first.
6427 The default value is @samp{tff}.
6429 The transformation is done by shifting the picture content up or down
6430 by one line, and filling the remaining line with appropriate picture content.
6431 This method is consistent with most broadcast field order converters.
6433 If the input video is not flagged as being interlaced, or it is already
6434 flagged as being of the required output field order, then this filter does
6435 not alter the incoming video.
6437 It is very useful when converting to or from PAL DV material,
6438 which is bottom field first.
6442 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
6445 @section fifo, afifo
6447 Buffer input images and send them when they are requested.
6449 It is mainly useful when auto-inserted by the libavfilter
6452 It does not take parameters.
6456 Find a rectangular object
6458 It accepts the following options:
6462 Filepath of the object image, needs to be in gray8.
6465 Detection threshold, default is 0.5.
6468 Number of mipmaps, default is 3.
6470 @item xmin, ymin, xmax, ymax
6471 Specifies the rectangle in which to search.
6474 @subsection Examples
6478 Generate a representative palette of a given video using @command{ffmpeg}:
6480 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6486 Cover a rectangular object
6488 It accepts the following options:
6492 Filepath of the optional cover image, needs to be in yuv420.
6497 It accepts the following values:
6500 cover it by the supplied image
6502 cover it by interpolating the surrounding pixels
6505 Default value is @var{blur}.
6508 @subsection Examples
6512 Generate a representative palette of a given video using @command{ffmpeg}:
6514 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6521 Convert the input video to one of the specified pixel formats.
6522 Libavfilter will try to pick one that is suitable as input to
6525 It accepts the following parameters:
6529 A '|'-separated list of pixel format names, such as
6530 "pix_fmts=yuv420p|monow|rgb24".
6534 @subsection Examples
6538 Convert the input video to the @var{yuv420p} format
6540 format=pix_fmts=yuv420p
6543 Convert the input video to any of the formats in the list
6545 format=pix_fmts=yuv420p|yuv444p|yuv410p
6552 Convert the video to specified constant frame rate by duplicating or dropping
6553 frames as necessary.
6555 It accepts the following parameters:
6559 The desired output frame rate. The default is @code{25}.
6564 Possible values are:
6567 zero round towards 0
6571 round towards -infinity
6573 round towards +infinity
6577 The default is @code{near}.
6580 Assume the first PTS should be the given value, in seconds. This allows for
6581 padding/trimming at the start of stream. By default, no assumption is made
6582 about the first frame's expected PTS, so no padding or trimming is done.
6583 For example, this could be set to 0 to pad the beginning with duplicates of
6584 the first frame if a video stream starts after the audio stream or to trim any
6585 frames with a negative PTS.
6589 Alternatively, the options can be specified as a flat string:
6590 @var{fps}[:@var{round}].
6592 See also the @ref{setpts} filter.
6594 @subsection Examples
6598 A typical usage in order to set the fps to 25:
6604 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
6606 fps=fps=film:round=near
6612 Pack two different video streams into a stereoscopic video, setting proper
6613 metadata on supported codecs. The two views should have the same size and
6614 framerate and processing will stop when the shorter video ends. Please note
6615 that you may conveniently adjust view properties with the @ref{scale} and
6618 It accepts the following parameters:
6622 The desired packing format. Supported values are:
6627 The views are next to each other (default).
6630 The views are on top of each other.
6633 The views are packed by line.
6636 The views are packed by column.
6639 The views are temporally interleaved.
6648 # Convert left and right views into a frame-sequential video
6649 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
6651 # Convert views into a side-by-side video with the same output resolution as the input
6652 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
6657 Change the frame rate by interpolating new video output frames from the source
6660 This filter is not designed to function correctly with interlaced media. If
6661 you wish to change the frame rate of interlaced media then you are required
6662 to deinterlace before this filter and re-interlace after this filter.
6664 A description of the accepted options follows.
6668 Specify the output frames per second. This option can also be specified
6669 as a value alone. The default is @code{50}.
6672 Specify the start of a range where the output frame will be created as a
6673 linear interpolation of two frames. The range is [@code{0}-@code{255}],
6674 the default is @code{15}.
6677 Specify the end of a range where the output frame will be created as a
6678 linear interpolation of two frames. The range is [@code{0}-@code{255}],
6679 the default is @code{240}.
6682 Specify the level at which a scene change is detected as a value between
6683 0 and 100 to indicate a new scene; a low value reflects a low
6684 probability for the current frame to introduce a new scene, while a higher
6685 value means the current frame is more likely to be one.
6686 The default is @code{7}.
6689 Specify flags influencing the filter process.
6691 Available value for @var{flags} is:
6694 @item scene_change_detect, scd
6695 Enable scene change detection using the value of the option @var{scene}.
6696 This flag is enabled by default.
6702 Select one frame every N-th frame.
6704 This filter accepts the following option:
6707 Select frame after every @code{step} frames.
6708 Allowed values are positive integers higher than 0. Default value is @code{1}.
6714 Apply a frei0r effect to the input video.
6716 To enable the compilation of this filter, you need to install the frei0r
6717 header and configure FFmpeg with @code{--enable-frei0r}.
6719 It accepts the following parameters:
6724 The name of the frei0r effect to load. If the environment variable
6725 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
6726 directories specified by the colon-separated list in @env{FREIOR_PATH}.
6727 Otherwise, the standard frei0r paths are searched, in this order:
6728 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
6729 @file{/usr/lib/frei0r-1/}.
6732 A '|'-separated list of parameters to pass to the frei0r effect.
6736 A frei0r effect parameter can be a boolean (its value is either
6737 "y" or "n"), a double, a color (specified as
6738 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
6739 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
6740 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
6741 @var{X} and @var{Y} are floating point numbers) and/or a string.
6743 The number and types of parameters depend on the loaded effect. If an
6744 effect parameter is not specified, the default value is set.
6746 @subsection Examples
6750 Apply the distort0r effect, setting the first two double parameters:
6752 frei0r=filter_name=distort0r:filter_params=0.5|0.01
6756 Apply the colordistance effect, taking a color as the first parameter:
6758 frei0r=colordistance:0.2/0.3/0.4
6759 frei0r=colordistance:violet
6760 frei0r=colordistance:0x112233
6764 Apply the perspective effect, specifying the top left and top right image
6767 frei0r=perspective:0.2/0.2|0.8/0.2
6771 For more information, see
6772 @url{http://frei0r.dyne.org}
6776 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
6778 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
6779 processing filter, one of them is performed once per block, not per pixel.
6780 This allows for much higher speed.
6782 The filter accepts the following options:
6786 Set quality. This option defines the number of levels for averaging. It accepts
6787 an integer in the range 4-5. Default value is @code{4}.
6790 Force a constant quantization parameter. It accepts an integer in range 0-63.
6791 If not set, the filter will use the QP from the video stream (if available).
6794 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
6795 more details but also more artifacts, while higher values make the image smoother
6796 but also blurrier. Default value is @code{0} − PSNR optimal.
6799 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
6800 option may cause flicker since the B-Frames have often larger QP. Default is
6801 @code{0} (not enabled).
6807 The filter accepts the following options:
6811 Set the luminance expression.
6813 Set the chrominance blue expression.
6815 Set the chrominance red expression.
6817 Set the alpha expression.
6819 Set the red expression.
6821 Set the green expression.
6823 Set the blue expression.
6826 The colorspace is selected according to the specified options. If one
6827 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
6828 options is specified, the filter will automatically select a YCbCr
6829 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
6830 @option{blue_expr} options is specified, it will select an RGB
6833 If one of the chrominance expression is not defined, it falls back on the other
6834 one. If no alpha expression is specified it will evaluate to opaque value.
6835 If none of chrominance expressions are specified, they will evaluate
6836 to the luminance expression.
6838 The expressions can use the following variables and functions:
6842 The sequential number of the filtered frame, starting from @code{0}.
6846 The coordinates of the current sample.
6850 The width and height of the image.
6854 Width and height scale depending on the currently filtered plane. It is the
6855 ratio between the corresponding luma plane number of pixels and the current
6856 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
6857 @code{0.5,0.5} for chroma planes.
6860 Time of the current frame, expressed in seconds.
6863 Return the value of the pixel at location (@var{x},@var{y}) of the current
6867 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
6871 Return the value of the pixel at location (@var{x},@var{y}) of the
6872 blue-difference chroma plane. Return 0 if there is no such plane.
6875 Return the value of the pixel at location (@var{x},@var{y}) of the
6876 red-difference chroma plane. Return 0 if there is no such plane.
6881 Return the value of the pixel at location (@var{x},@var{y}) of the
6882 red/green/blue component. Return 0 if there is no such component.
6885 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
6886 plane. Return 0 if there is no such plane.
6889 For functions, if @var{x} and @var{y} are outside the area, the value will be
6890 automatically clipped to the closer edge.
6892 @subsection Examples
6896 Flip the image horizontally:
6902 Generate a bidimensional sine wave, with angle @code{PI/3} and a
6903 wavelength of 100 pixels:
6905 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
6909 Generate a fancy enigmatic moving light:
6911 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
6915 Generate a quick emboss effect:
6917 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
6921 Modify RGB components depending on pixel position:
6923 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
6927 Create a radial gradient that is the same size as the input (also see
6928 the @ref{vignette} filter):
6930 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
6934 Create a linear gradient to use as a mask for another filter, then
6935 compose with @ref{overlay}. In this example the video will gradually
6936 become more blurry from the top to the bottom of the y-axis as defined
6937 by the linear gradient:
6939 ffmpeg -i input.mp4 -filter_complex "geq=lum=255*(Y/H),format=gray[grad];[0:v]boxblur=4[blur];[blur][grad]alphamerge[alpha];[0:v][alpha]overlay" output.mp4
6945 Fix the banding artifacts that are sometimes introduced into nearly flat
6946 regions by truncation to 8bit color depth.
6947 Interpolate the gradients that should go where the bands are, and
6950 It is designed for playback only. Do not use it prior to
6951 lossy compression, because compression tends to lose the dither and
6952 bring back the bands.
6954 It accepts the following parameters:
6959 The maximum amount by which the filter will change any one pixel. This is also
6960 the threshold for detecting nearly flat regions. Acceptable values range from
6961 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
6965 The neighborhood to fit the gradient to. A larger radius makes for smoother
6966 gradients, but also prevents the filter from modifying the pixels near detailed
6967 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
6968 values will be clipped to the valid range.
6972 Alternatively, the options can be specified as a flat string:
6973 @var{strength}[:@var{radius}]
6975 @subsection Examples
6979 Apply the filter with a @code{3.5} strength and radius of @code{8}:
6985 Specify radius, omitting the strength (which will fall-back to the default
6996 Apply a Hald CLUT to a video stream.
6998 First input is the video stream to process, and second one is the Hald CLUT.
6999 The Hald CLUT input can be a simple picture or a complete video stream.
7001 The filter accepts the following options:
7005 Force termination when the shortest input terminates. Default is @code{0}.
7007 Continue applying the last CLUT after the end of the stream. A value of
7008 @code{0} disable the filter after the last frame of the CLUT is reached.
7009 Default is @code{1}.
7012 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
7013 filters share the same internals).
7015 More information about the Hald CLUT can be found on Eskil Steenberg's website
7016 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
7018 @subsection Workflow examples
7020 @subsubsection Hald CLUT video stream
7022 Generate an identity Hald CLUT stream altered with various effects:
7024 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
7027 Note: make sure you use a lossless codec.
7029 Then use it with @code{haldclut} to apply it on some random stream:
7031 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
7034 The Hald CLUT will be applied to the 10 first seconds (duration of
7035 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
7036 to the remaining frames of the @code{mandelbrot} stream.
7038 @subsubsection Hald CLUT with preview
7040 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
7041 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
7042 biggest possible square starting at the top left of the picture. The remaining
7043 padding pixels (bottom or right) will be ignored. This area can be used to add
7044 a preview of the Hald CLUT.
7046 Typically, the following generated Hald CLUT will be supported by the
7047 @code{haldclut} filter:
7050 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
7051 pad=iw+320 [padded_clut];
7052 smptebars=s=320x256, split [a][b];
7053 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
7054 [main][b] overlay=W-320" -frames:v 1 clut.png
7057 It contains the original and a preview of the effect of the CLUT: SMPTE color
7058 bars are displayed on the right-top, and below the same color bars processed by
7061 Then, the effect of this Hald CLUT can be visualized with:
7063 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
7068 Flip the input video horizontally.
7070 For example, to horizontally flip the input video with @command{ffmpeg}:
7072 ffmpeg -i in.avi -vf "hflip" out.avi
7076 This filter applies a global color histogram equalization on a
7079 It can be used to correct video that has a compressed range of pixel
7080 intensities. The filter redistributes the pixel intensities to
7081 equalize their distribution across the intensity range. It may be
7082 viewed as an "automatically adjusting contrast filter". This filter is
7083 useful only for correcting degraded or poorly captured source
7086 The filter accepts the following options:
7090 Determine the amount of equalization to be applied. As the strength
7091 is reduced, the distribution of pixel intensities more-and-more
7092 approaches that of the input frame. The value must be a float number
7093 in the range [0,1] and defaults to 0.200.
7096 Set the maximum intensity that can generated and scale the output
7097 values appropriately. The strength should be set as desired and then
7098 the intensity can be limited if needed to avoid washing-out. The value
7099 must be a float number in the range [0,1] and defaults to 0.210.
7102 Set the antibanding level. If enabled the filter will randomly vary
7103 the luminance of output pixels by a small amount to avoid banding of
7104 the histogram. Possible values are @code{none}, @code{weak} or
7105 @code{strong}. It defaults to @code{none}.
7110 Compute and draw a color distribution histogram for the input video.
7112 The computed histogram is a representation of the color component
7113 distribution in an image.
7115 The filter accepts the following options:
7121 It accepts the following values:
7124 Standard histogram that displays the color components distribution in an
7125 image. Displays color graph for each color component. Shows distribution of
7126 the Y, U, V, A or R, G, B components, depending on input format, in the
7127 current frame. Below each graph a color component scale meter is shown.
7130 Displays chroma values (U/V color placement) in a two dimensional
7131 graph (which is called a vectorscope). The brighter a pixel in the
7132 vectorscope, the more pixels of the input frame correspond to that pixel
7133 (i.e., more pixels have this chroma value). The V component is displayed on
7134 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
7135 side being V = 255. The U component is displayed on the vertical (Y) axis,
7136 with the top representing U = 0 and the bottom representing U = 255.
7138 The position of a white pixel in the graph corresponds to the chroma value of
7139 a pixel of the input clip. The graph can therefore be used to read the hue
7140 (color flavor) and the saturation (the dominance of the hue in the color). As
7141 the hue of a color changes, it moves around the square. At the center of the
7142 square the saturation is zero, which means that the corresponding pixel has no
7143 color. If the amount of a specific color is increased (while leaving the other
7144 colors unchanged) the saturation increases, and the indicator moves towards
7145 the edge of the square.
7148 Chroma values in vectorscope, similar as @code{color} but actual chroma values
7152 Per row/column color component graph. In row mode, the graph on the left side
7153 represents color component value 0 and the right side represents value = 255.
7154 In column mode, the top side represents color component value = 0 and bottom
7155 side represents value = 255.
7157 Default value is @code{levels}.
7160 Set height of level in @code{levels}. Default value is @code{200}.
7161 Allowed range is [50, 2048].
7164 Set height of color scale in @code{levels}. Default value is @code{12}.
7165 Allowed range is [0, 40].
7168 Set step for @code{waveform} mode. Smaller values are useful to find out how
7169 many values of the same luminance are distributed across input rows/columns.
7170 Default value is @code{10}. Allowed range is [1, 255].
7173 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
7174 Default is @code{row}.
7176 @item waveform_mirror
7177 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
7178 means mirrored. In mirrored mode, higher values will be represented on the left
7179 side for @code{row} mode and at the top for @code{column} mode. Default is
7180 @code{0} (unmirrored).
7183 Set display mode for @code{waveform} and @code{levels}.
7184 It accepts the following values:
7187 Display separate graph for the color components side by side in
7188 @code{row} waveform mode or one below the other in @code{column} waveform mode
7189 for @code{waveform} histogram mode. For @code{levels} histogram mode,
7190 per color component graphs are placed below each other.
7192 Using this display mode in @code{waveform} histogram mode makes it easy to
7193 spot color casts in the highlights and shadows of an image, by comparing the
7194 contours of the top and the bottom graphs of each waveform. Since whites,
7195 grays, and blacks are characterized by exactly equal amounts of red, green,
7196 and blue, neutral areas of the picture should display three waveforms of
7197 roughly equal width/height. If not, the correction is easy to perform by
7198 making level adjustments the three waveforms.
7201 Presents information identical to that in the @code{parade}, except
7202 that the graphs representing color components are superimposed directly
7205 This display mode in @code{waveform} histogram mode makes it easier to spot
7206 relative differences or similarities in overlapping areas of the color
7207 components that are supposed to be identical, such as neutral whites, grays,
7210 Default is @code{parade}.
7213 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
7214 Default is @code{linear}.
7217 Set what color components to display for mode @code{levels}.
7218 Default is @code{7}.
7221 @subsection Examples
7226 Calculate and draw histogram:
7228 ffplay -i input -vf histogram
7236 This is a high precision/quality 3d denoise filter. It aims to reduce
7237 image noise, producing smooth images and making still images really
7238 still. It should enhance compressibility.
7240 It accepts the following optional parameters:
7244 A non-negative floating point number which specifies spatial luma strength.
7247 @item chroma_spatial
7248 A non-negative floating point number which specifies spatial chroma strength.
7249 It defaults to 3.0*@var{luma_spatial}/4.0.
7252 A floating point number which specifies luma temporal strength. It defaults to
7253 6.0*@var{luma_spatial}/4.0.
7256 A floating point number which specifies chroma temporal strength. It defaults to
7257 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
7262 Apply a high-quality magnification filter designed for pixel art. This filter
7263 was originally created by Maxim Stepin.
7265 It accepts the following option:
7269 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
7270 @code{hq3x} and @code{4} for @code{hq4x}.
7271 Default is @code{3}.
7275 Stack input videos horizontally.
7277 All streams must be of same pixel format and of same height.
7279 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
7280 to create same output.
7282 The filter accept the following option:
7286 Set number of input streams. Default is 2.
7289 If set to 1, force the output to terminate when the shortest input
7290 terminates. Default value is 0.
7295 Modify the hue and/or the saturation of the input.
7297 It accepts the following parameters:
7301 Specify the hue angle as a number of degrees. It accepts an expression,
7302 and defaults to "0".
7305 Specify the saturation in the [-10,10] range. It accepts an expression and
7309 Specify the hue angle as a number of radians. It accepts an
7310 expression, and defaults to "0".
7313 Specify the brightness in the [-10,10] range. It accepts an expression and
7317 @option{h} and @option{H} are mutually exclusive, and can't be
7318 specified at the same time.
7320 The @option{b}, @option{h}, @option{H} and @option{s} option values are
7321 expressions containing the following constants:
7325 frame count of the input frame starting from 0
7328 presentation timestamp of the input frame expressed in time base units
7331 frame rate of the input video, NAN if the input frame rate is unknown
7334 timestamp expressed in seconds, NAN if the input timestamp is unknown
7337 time base of the input video
7340 @subsection Examples
7344 Set the hue to 90 degrees and the saturation to 1.0:
7350 Same command but expressing the hue in radians:
7356 Rotate hue and make the saturation swing between 0
7357 and 2 over a period of 1 second:
7359 hue="H=2*PI*t: s=sin(2*PI*t)+1"
7363 Apply a 3 seconds saturation fade-in effect starting at 0:
7368 The general fade-in expression can be written as:
7370 hue="s=min(0\, max((t-START)/DURATION\, 1))"
7374 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
7376 hue="s=max(0\, min(1\, (8-t)/3))"
7379 The general fade-out expression can be written as:
7381 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
7386 @subsection Commands
7388 This filter supports the following commands:
7394 Modify the hue and/or the saturation and/or brightness of the input video.
7395 The command accepts the same syntax of the corresponding option.
7397 If the specified expression is not valid, it is kept at its current
7403 Detect video interlacing type.
7405 This filter tries to detect if the input frames as interlaced, progressive,
7406 top or bottom field first. It will also try and detect fields that are
7407 repeated between adjacent frames (a sign of telecine).
7409 Single frame detection considers only immediately adjacent frames when classifying each frame.
7410 Multiple frame detection incorporates the classification history of previous frames.
7412 The filter will log these metadata values:
7415 @item single.current_frame
7416 Detected type of current frame using single-frame detection. One of:
7417 ``tff'' (top field first), ``bff'' (bottom field first),
7418 ``progressive'', or ``undetermined''
7421 Cumulative number of frames detected as top field first using single-frame detection.
7424 Cumulative number of frames detected as top field first using multiple-frame detection.
7427 Cumulative number of frames detected as bottom field first using single-frame detection.
7429 @item multiple.current_frame
7430 Detected type of current frame using multiple-frame detection. One of:
7431 ``tff'' (top field first), ``bff'' (bottom field first),
7432 ``progressive'', or ``undetermined''
7435 Cumulative number of frames detected as bottom field first using multiple-frame detection.
7437 @item single.progressive
7438 Cumulative number of frames detected as progressive using single-frame detection.
7440 @item multiple.progressive
7441 Cumulative number of frames detected as progressive using multiple-frame detection.
7443 @item single.undetermined
7444 Cumulative number of frames that could not be classified using single-frame detection.
7446 @item multiple.undetermined
7447 Cumulative number of frames that could not be classified using multiple-frame detection.
7449 @item repeated.current_frame
7450 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
7452 @item repeated.neither
7453 Cumulative number of frames with no repeated field.
7456 Cumulative number of frames with the top field repeated from the previous frame's top field.
7458 @item repeated.bottom
7459 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
7462 The filter accepts the following options:
7466 Set interlacing threshold.
7468 Set progressive threshold.
7470 Threshold for repeated field detection.
7472 Number of frames after which a given frame's contribution to the
7473 statistics is halved (i.e., it contributes only 0.5 to it's
7474 classification). The default of 0 means that all frames seen are given
7475 full weight of 1.0 forever.
7476 @item analyze_interlaced_flag
7477 When this is not 0 then idet will use the specified number of frames to determine
7478 if the interlaced flag is accurate, it will not count undetermined frames.
7479 If the flag is found to be accurate it will be used without any further
7480 computations, if it is found to be inaccurate it will be cleared without any
7481 further computations. This allows inserting the idet filter as a low computational
7482 method to clean up the interlaced flag
7487 Deinterleave or interleave fields.
7489 This filter allows one to process interlaced images fields without
7490 deinterlacing them. Deinterleaving splits the input frame into 2
7491 fields (so called half pictures). Odd lines are moved to the top
7492 half of the output image, even lines to the bottom half.
7493 You can process (filter) them independently and then re-interleave them.
7495 The filter accepts the following options:
7499 @item chroma_mode, c
7501 Available values for @var{luma_mode}, @var{chroma_mode} and
7502 @var{alpha_mode} are:
7508 @item deinterleave, d
7509 Deinterleave fields, placing one above the other.
7512 Interleave fields. Reverse the effect of deinterleaving.
7514 Default value is @code{none}.
7517 @item chroma_swap, cs
7518 @item alpha_swap, as
7519 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
7524 Apply inflate effect to the video.
7526 This filter replaces the pixel by the local(3x3) average by taking into account
7527 only values higher than the pixel.
7529 It accepts the following options:
7536 Limit the maximum change for each plane, default is 65535.
7537 If 0, plane will remain unchanged.
7542 Simple interlacing filter from progressive contents. This interleaves upper (or
7543 lower) lines from odd frames with lower (or upper) lines from even frames,
7544 halving the frame rate and preserving image height.
7547 Original Original New Frame
7548 Frame 'j' Frame 'j+1' (tff)
7549 ========== =========== ==================
7550 Line 0 --------------------> Frame 'j' Line 0
7551 Line 1 Line 1 ----> Frame 'j+1' Line 1
7552 Line 2 ---------------------> Frame 'j' Line 2
7553 Line 3 Line 3 ----> Frame 'j+1' Line 3
7555 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
7558 It accepts the following optional parameters:
7562 This determines whether the interlaced frame is taken from the even
7563 (tff - default) or odd (bff) lines of the progressive frame.
7566 Enable (default) or disable the vertical lowpass filter to avoid twitter
7567 interlacing and reduce moire patterns.
7572 Deinterlace input video by applying Donald Graft's adaptive kernel
7573 deinterling. Work on interlaced parts of a video to produce
7576 The description of the accepted parameters follows.
7580 Set the threshold which affects the filter's tolerance when
7581 determining if a pixel line must be processed. It must be an integer
7582 in the range [0,255] and defaults to 10. A value of 0 will result in
7583 applying the process on every pixels.
7586 Paint pixels exceeding the threshold value to white if set to 1.
7590 Set the fields order. Swap fields if set to 1, leave fields alone if
7594 Enable additional sharpening if set to 1. Default is 0.
7597 Enable twoway sharpening if set to 1. Default is 0.
7600 @subsection Examples
7604 Apply default values:
7606 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
7610 Enable additional sharpening:
7616 Paint processed pixels in white:
7622 @section lenscorrection
7624 Correct radial lens distortion
7626 This filter can be used to correct for radial distortion as can result from the use
7627 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
7628 one can use tools available for example as part of opencv or simply trial-and-error.
7629 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
7630 and extract the k1 and k2 coefficients from the resulting matrix.
7632 Note that effectively the same filter is available in the open-source tools Krita and
7633 Digikam from the KDE project.
7635 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
7636 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
7637 brightness distribution, so you may want to use both filters together in certain
7638 cases, though you will have to take care of ordering, i.e. whether vignetting should
7639 be applied before or after lens correction.
7643 The filter accepts the following options:
7647 Relative x-coordinate of the focal point of the image, and thereby the center of the
7648 distortion. This value has a range [0,1] and is expressed as fractions of the image
7651 Relative y-coordinate of the focal point of the image, and thereby the center of the
7652 distortion. This value has a range [0,1] and is expressed as fractions of the image
7655 Coefficient of the quadratic correction term. 0.5 means no correction.
7657 Coefficient of the double quadratic correction term. 0.5 means no correction.
7660 The formula that generates the correction is:
7662 @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)
7664 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
7665 distances from the focal point in the source and target images, respectively.
7670 Apply a 3D LUT to an input video.
7672 The filter accepts the following options:
7676 Set the 3D LUT file name.
7678 Currently supported formats:
7690 Select interpolation mode.
7692 Available values are:
7696 Use values from the nearest defined point.
7698 Interpolate values using the 8 points defining a cube.
7700 Interpolate values using a tetrahedron.
7704 @section lut, lutrgb, lutyuv
7706 Compute a look-up table for binding each pixel component input value
7707 to an output value, and apply it to the input video.
7709 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
7710 to an RGB input video.
7712 These filters accept the following parameters:
7715 set first pixel component expression
7717 set second pixel component expression
7719 set third pixel component expression
7721 set fourth pixel component expression, corresponds to the alpha component
7724 set red component expression
7726 set green component expression
7728 set blue component expression
7730 alpha component expression
7733 set Y/luminance component expression
7735 set U/Cb component expression
7737 set V/Cr component expression
7740 Each of them specifies the expression to use for computing the lookup table for
7741 the corresponding pixel component values.
7743 The exact component associated to each of the @var{c*} options depends on the
7746 The @var{lut} filter requires either YUV or RGB pixel formats in input,
7747 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
7749 The expressions can contain the following constants and functions:
7754 The input width and height.
7757 The input value for the pixel component.
7760 The input value, clipped to the @var{minval}-@var{maxval} range.
7763 The maximum value for the pixel component.
7766 The minimum value for the pixel component.
7769 The negated value for the pixel component value, clipped to the
7770 @var{minval}-@var{maxval} range; it corresponds to the expression
7771 "maxval-clipval+minval".
7774 The computed value in @var{val}, clipped to the
7775 @var{minval}-@var{maxval} range.
7777 @item gammaval(gamma)
7778 The computed gamma correction value of the pixel component value,
7779 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
7781 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
7785 All expressions default to "val".
7787 @subsection Examples
7793 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
7794 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
7797 The above is the same as:
7799 lutrgb="r=negval:g=negval:b=negval"
7800 lutyuv="y=negval:u=negval:v=negval"
7810 Remove chroma components, turning the video into a graytone image:
7812 lutyuv="u=128:v=128"
7816 Apply a luma burning effect:
7822 Remove green and blue components:
7828 Set a constant alpha channel value on input:
7830 format=rgba,lutrgb=a="maxval-minval/2"
7834 Correct luminance gamma by a factor of 0.5:
7836 lutyuv=y=gammaval(0.5)
7840 Discard least significant bits of luma:
7842 lutyuv=y='bitand(val, 128+64+32)'
7846 @section maskedmerge
7848 Merge the first input stream with the second input stream using per pixel
7849 weights in the third input stream.
7851 A value of 0 in the third stream pixel component means that pixel component
7852 from first stream is returned unchanged, while maximum value (eg. 255 for
7853 8-bit videos) means that pixel component from second stream is returned
7854 unchanged. Intermediate values define the amount of merging between both
7855 input stream's pixel components.
7857 This filter accepts the following options:
7860 Set which planes will be processed as bitmap, unprocessed planes will be
7861 copied from first stream.
7862 By default value 0xf, all planes will be processed.
7867 Apply motion-compensation deinterlacing.
7869 It needs one field per frame as input and must thus be used together
7870 with yadif=1/3 or equivalent.
7872 This filter accepts the following options:
7875 Set the deinterlacing mode.
7877 It accepts one of the following values:
7882 use iterative motion estimation
7884 like @samp{slow}, but use multiple reference frames.
7886 Default value is @samp{fast}.
7889 Set the picture field parity assumed for the input video. It must be
7890 one of the following values:
7894 assume top field first
7896 assume bottom field first
7899 Default value is @samp{bff}.
7902 Set per-block quantization parameter (QP) used by the internal
7905 Higher values should result in a smoother motion vector field but less
7906 optimal individual vectors. Default value is 1.
7909 @section mergeplanes
7911 Merge color channel components from several video streams.
7913 The filter accepts up to 4 input streams, and merge selected input
7914 planes to the output video.
7916 This filter accepts the following options:
7919 Set input to output plane mapping. Default is @code{0}.
7921 The mappings is specified as a bitmap. It should be specified as a
7922 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
7923 mapping for the first plane of the output stream. 'A' sets the number of
7924 the input stream to use (from 0 to 3), and 'a' the plane number of the
7925 corresponding input to use (from 0 to 3). The rest of the mappings is
7926 similar, 'Bb' describes the mapping for the output stream second
7927 plane, 'Cc' describes the mapping for the output stream third plane and
7928 'Dd' describes the mapping for the output stream fourth plane.
7931 Set output pixel format. Default is @code{yuva444p}.
7934 @subsection Examples
7938 Merge three gray video streams of same width and height into single video stream:
7940 [a0][a1][a2]mergeplanes=0x001020:yuv444p
7944 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
7946 [a0][a1]mergeplanes=0x00010210:yuva444p
7950 Swap Y and A plane in yuva444p stream:
7952 format=yuva444p,mergeplanes=0x03010200:yuva444p
7956 Swap U and V plane in yuv420p stream:
7958 format=yuv420p,mergeplanes=0x000201:yuv420p
7962 Cast a rgb24 clip to yuv444p:
7964 format=rgb24,mergeplanes=0x000102:yuv444p
7970 Drop frames that do not differ greatly from the previous frame in
7971 order to reduce frame rate.
7973 The main use of this filter is for very-low-bitrate encoding
7974 (e.g. streaming over dialup modem), but it could in theory be used for
7975 fixing movies that were inverse-telecined incorrectly.
7977 A description of the accepted options follows.
7981 Set the maximum number of consecutive frames which can be dropped (if
7982 positive), or the minimum interval between dropped frames (if
7983 negative). If the value is 0, the frame is dropped unregarding the
7984 number of previous sequentially dropped frames.
7991 Set the dropping threshold values.
7993 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
7994 represent actual pixel value differences, so a threshold of 64
7995 corresponds to 1 unit of difference for each pixel, or the same spread
7996 out differently over the block.
7998 A frame is a candidate for dropping if no 8x8 blocks differ by more
7999 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
8000 meaning the whole image) differ by more than a threshold of @option{lo}.
8002 Default value for @option{hi} is 64*12, default value for @option{lo} is
8003 64*5, and default value for @option{frac} is 0.33.
8011 It accepts an integer in input; if non-zero it negates the
8012 alpha component (if available). The default value in input is 0.
8016 Force libavfilter not to use any of the specified pixel formats for the
8017 input to the next filter.
8019 It accepts the following parameters:
8023 A '|'-separated list of pixel format names, such as
8024 apix_fmts=yuv420p|monow|rgb24".
8028 @subsection Examples
8032 Force libavfilter to use a format different from @var{yuv420p} for the
8033 input to the vflip filter:
8035 noformat=pix_fmts=yuv420p,vflip
8039 Convert the input video to any of the formats not contained in the list:
8041 noformat=yuv420p|yuv444p|yuv410p
8047 Add noise on video input frame.
8049 The filter accepts the following options:
8057 Set noise seed for specific pixel component or all pixel components in case
8058 of @var{all_seed}. Default value is @code{123457}.
8060 @item all_strength, alls
8061 @item c0_strength, c0s
8062 @item c1_strength, c1s
8063 @item c2_strength, c2s
8064 @item c3_strength, c3s
8065 Set noise strength for specific pixel component or all pixel components in case
8066 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
8068 @item all_flags, allf
8073 Set pixel component flags or set flags for all components if @var{all_flags}.
8074 Available values for component flags are:
8077 averaged temporal noise (smoother)
8079 mix random noise with a (semi)regular pattern
8081 temporal noise (noise pattern changes between frames)
8083 uniform noise (gaussian otherwise)
8087 @subsection Examples
8089 Add temporal and uniform noise to input video:
8091 noise=alls=20:allf=t+u
8096 Pass the video source unchanged to the output.
8099 Optical Character Recognition
8101 This filter uses Tesseract for optical character recognition.
8103 It accepts the following options:
8107 Set datapath to tesseract data. Default is to use whatever was
8108 set at installation.
8111 Set language, default is "eng".
8114 Set character whitelist.
8117 Set character blacklist.
8120 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
8124 Apply a video transform using libopencv.
8126 To enable this filter, install the libopencv library and headers and
8127 configure FFmpeg with @code{--enable-libopencv}.
8129 It accepts the following parameters:
8134 The name of the libopencv filter to apply.
8137 The parameters to pass to the libopencv filter. If not specified, the default
8142 Refer to the official libopencv documentation for more precise
8144 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
8146 Several libopencv filters are supported; see the following subsections.
8151 Dilate an image by using a specific structuring element.
8152 It corresponds to the libopencv function @code{cvDilate}.
8154 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
8156 @var{struct_el} represents a structuring element, and has the syntax:
8157 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
8159 @var{cols} and @var{rows} represent the number of columns and rows of
8160 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
8161 point, and @var{shape} the shape for the structuring element. @var{shape}
8162 must be "rect", "cross", "ellipse", or "custom".
8164 If the value for @var{shape} is "custom", it must be followed by a
8165 string of the form "=@var{filename}". The file with name
8166 @var{filename} is assumed to represent a binary image, with each
8167 printable character corresponding to a bright pixel. When a custom
8168 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
8169 or columns and rows of the read file are assumed instead.
8171 The default value for @var{struct_el} is "3x3+0x0/rect".
8173 @var{nb_iterations} specifies the number of times the transform is
8174 applied to the image, and defaults to 1.
8178 # Use the default values
8181 # Dilate using a structuring element with a 5x5 cross, iterating two times
8182 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
8184 # Read the shape from the file diamond.shape, iterating two times.
8185 # The file diamond.shape may contain a pattern of characters like this
8191 # The specified columns and rows are ignored
8192 # but the anchor point coordinates are not
8193 ocv=dilate:0x0+2x2/custom=diamond.shape|2
8198 Erode an image by using a specific structuring element.
8199 It corresponds to the libopencv function @code{cvErode}.
8201 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
8202 with the same syntax and semantics as the @ref{dilate} filter.
8206 Smooth the input video.
8208 The filter takes the following parameters:
8209 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
8211 @var{type} is the type of smooth filter to apply, and must be one of
8212 the following values: "blur", "blur_no_scale", "median", "gaussian",
8213 or "bilateral". The default value is "gaussian".
8215 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
8216 depend on the smooth type. @var{param1} and
8217 @var{param2} accept integer positive values or 0. @var{param3} and
8218 @var{param4} accept floating point values.
8220 The default value for @var{param1} is 3. The default value for the
8221 other parameters is 0.
8223 These parameters correspond to the parameters assigned to the
8224 libopencv function @code{cvSmooth}.
8229 Overlay one video on top of another.
8231 It takes two inputs and has one output. The first input is the "main"
8232 video on which the second input is overlaid.
8234 It accepts the following parameters:
8236 A description of the accepted options follows.
8241 Set the expression for the x and y coordinates of the overlaid video
8242 on the main video. Default value is "0" for both expressions. In case
8243 the expression is invalid, it is set to a huge value (meaning that the
8244 overlay will not be displayed within the output visible area).
8247 The action to take when EOF is encountered on the secondary input; it accepts
8248 one of the following values:
8252 Repeat the last frame (the default).
8256 Pass the main input through.
8260 Set when the expressions for @option{x}, and @option{y} are evaluated.
8262 It accepts the following values:
8265 only evaluate expressions once during the filter initialization or
8266 when a command is processed
8269 evaluate expressions for each incoming frame
8272 Default value is @samp{frame}.
8275 If set to 1, force the output to terminate when the shortest input
8276 terminates. Default value is 0.
8279 Set the format for the output video.
8281 It accepts the following values:
8296 Default value is @samp{yuv420}.
8298 @item rgb @emph{(deprecated)}
8299 If set to 1, force the filter to accept inputs in the RGB
8300 color space. Default value is 0. This option is deprecated, use
8301 @option{format} instead.
8304 If set to 1, force the filter to draw the last overlay frame over the
8305 main input until the end of the stream. A value of 0 disables this
8306 behavior. Default value is 1.
8309 The @option{x}, and @option{y} expressions can contain the following
8315 The main input width and height.
8319 The overlay input width and height.
8323 The computed values for @var{x} and @var{y}. They are evaluated for
8328 horizontal and vertical chroma subsample values of the output
8329 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
8333 the number of input frame, starting from 0
8336 the position in the file of the input frame, NAN if unknown
8339 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
8343 Note that the @var{n}, @var{pos}, @var{t} variables are available only
8344 when evaluation is done @emph{per frame}, and will evaluate to NAN
8345 when @option{eval} is set to @samp{init}.
8347 Be aware that frames are taken from each input video in timestamp
8348 order, hence, if their initial timestamps differ, it is a good idea
8349 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
8350 have them begin in the same zero timestamp, as the example for
8351 the @var{movie} filter does.
8353 You can chain together more overlays but you should test the
8354 efficiency of such approach.
8356 @subsection Commands
8358 This filter supports the following commands:
8362 Modify the x and y of the overlay input.
8363 The command accepts the same syntax of the corresponding option.
8365 If the specified expression is not valid, it is kept at its current
8369 @subsection Examples
8373 Draw the overlay at 10 pixels from the bottom right corner of the main
8376 overlay=main_w-overlay_w-10:main_h-overlay_h-10
8379 Using named options the example above becomes:
8381 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
8385 Insert a transparent PNG logo in the bottom left corner of the input,
8386 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
8388 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
8392 Insert 2 different transparent PNG logos (second logo on bottom
8393 right corner) using the @command{ffmpeg} tool:
8395 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
8399 Add a transparent color layer on top of the main video; @code{WxH}
8400 must specify the size of the main input to the overlay filter:
8402 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
8406 Play an original video and a filtered version (here with the deshake
8407 filter) side by side using the @command{ffplay} tool:
8409 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
8412 The above command is the same as:
8414 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
8418 Make a sliding overlay appearing from the left to the right top part of the
8419 screen starting since time 2:
8421 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
8425 Compose output by putting two input videos side to side:
8427 ffmpeg -i left.avi -i right.avi -filter_complex "
8428 nullsrc=size=200x100 [background];
8429 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
8430 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
8431 [background][left] overlay=shortest=1 [background+left];
8432 [background+left][right] overlay=shortest=1:x=100 [left+right]
8437 Mask 10-20 seconds of a video by applying the delogo filter to a section
8439 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
8440 -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]'
8445 Chain several overlays in cascade:
8447 nullsrc=s=200x200 [bg];
8448 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
8449 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
8450 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
8451 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
8452 [in3] null, [mid2] overlay=100:100 [out0]
8459 Apply Overcomplete Wavelet denoiser.
8461 The filter accepts the following options:
8467 Larger depth values will denoise lower frequency components more, but
8468 slow down filtering.
8470 Must be an int in the range 8-16, default is @code{8}.
8472 @item luma_strength, ls
8475 Must be a double value in the range 0-1000, default is @code{1.0}.
8477 @item chroma_strength, cs
8478 Set chroma strength.
8480 Must be a double value in the range 0-1000, default is @code{1.0}.
8486 Add paddings to the input image, and place the original input at the
8487 provided @var{x}, @var{y} coordinates.
8489 It accepts the following parameters:
8494 Specify an expression for the size of the output image with the
8495 paddings added. If the value for @var{width} or @var{height} is 0, the
8496 corresponding input size is used for the output.
8498 The @var{width} expression can reference the value set by the
8499 @var{height} expression, and vice versa.
8501 The default value of @var{width} and @var{height} is 0.
8505 Specify the offsets to place the input image at within the padded area,
8506 with respect to the top/left border of the output image.
8508 The @var{x} expression can reference the value set by the @var{y}
8509 expression, and vice versa.
8511 The default value of @var{x} and @var{y} is 0.
8514 Specify the color of the padded area. For the syntax of this option,
8515 check the "Color" section in the ffmpeg-utils manual.
8517 The default value of @var{color} is "black".
8520 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
8521 options are expressions containing the following constants:
8526 The input video width and height.
8530 These are the same as @var{in_w} and @var{in_h}.
8534 The output width and height (the size of the padded area), as
8535 specified by the @var{width} and @var{height} expressions.
8539 These are the same as @var{out_w} and @var{out_h}.
8543 The x and y offsets as specified by the @var{x} and @var{y}
8544 expressions, or NAN if not yet specified.
8547 same as @var{iw} / @var{ih}
8550 input sample aspect ratio
8553 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
8557 The horizontal and vertical chroma subsample values. For example for the
8558 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8561 @subsection Examples
8565 Add paddings with the color "violet" to the input video. The output video
8566 size is 640x480, and the top-left corner of the input video is placed at
8569 pad=640:480:0:40:violet
8572 The example above is equivalent to the following command:
8574 pad=width=640:height=480:x=0:y=40:color=violet
8578 Pad the input to get an output with dimensions increased by 3/2,
8579 and put the input video at the center of the padded area:
8581 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
8585 Pad the input to get a squared output with size equal to the maximum
8586 value between the input width and height, and put the input video at
8587 the center of the padded area:
8589 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
8593 Pad the input to get a final w/h ratio of 16:9:
8595 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
8599 In case of anamorphic video, in order to set the output display aspect
8600 correctly, it is necessary to use @var{sar} in the expression,
8601 according to the relation:
8603 (ih * X / ih) * sar = output_dar
8604 X = output_dar / sar
8607 Thus the previous example needs to be modified to:
8609 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
8613 Double the output size and put the input video in the bottom-right
8614 corner of the output padded area:
8616 pad="2*iw:2*ih:ow-iw:oh-ih"
8623 Generate one palette for a whole video stream.
8625 It accepts the following options:
8629 Set the maximum number of colors to quantize in the palette.
8630 Note: the palette will still contain 256 colors; the unused palette entries
8633 @item reserve_transparent
8634 Create a palette of 255 colors maximum and reserve the last one for
8635 transparency. Reserving the transparency color is useful for GIF optimization.
8636 If not set, the maximum of colors in the palette will be 256. You probably want
8637 to disable this option for a standalone image.
8641 Set statistics mode.
8643 It accepts the following values:
8646 Compute full frame histograms.
8648 Compute histograms only for the part that differs from previous frame. This
8649 might be relevant to give more importance to the moving part of your input if
8650 the background is static.
8653 Default value is @var{full}.
8656 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
8657 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
8658 color quantization of the palette. This information is also visible at
8659 @var{info} logging level.
8661 @subsection Examples
8665 Generate a representative palette of a given video using @command{ffmpeg}:
8667 ffmpeg -i input.mkv -vf palettegen palette.png
8673 Use a palette to downsample an input video stream.
8675 The filter takes two inputs: one video stream and a palette. The palette must
8676 be a 256 pixels image.
8678 It accepts the following options:
8682 Select dithering mode. Available algorithms are:
8685 Ordered 8x8 bayer dithering (deterministic)
8687 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
8688 Note: this dithering is sometimes considered "wrong" and is included as a
8690 @item floyd_steinberg
8691 Floyd and Steingberg dithering (error diffusion)
8693 Frankie Sierra dithering v2 (error diffusion)
8695 Frankie Sierra dithering v2 "Lite" (error diffusion)
8698 Default is @var{sierra2_4a}.
8701 When @var{bayer} dithering is selected, this option defines the scale of the
8702 pattern (how much the crosshatch pattern is visible). A low value means more
8703 visible pattern for less banding, and higher value means less visible pattern
8704 at the cost of more banding.
8706 The option must be an integer value in the range [0,5]. Default is @var{2}.
8709 If set, define the zone to process
8713 Only the changing rectangle will be reprocessed. This is similar to GIF
8714 cropping/offsetting compression mechanism. This option can be useful for speed
8715 if only a part of the image is changing, and has use cases such as limiting the
8716 scope of the error diffusal @option{dither} to the rectangle that bounds the
8717 moving scene (it leads to more deterministic output if the scene doesn't change
8718 much, and as a result less moving noise and better GIF compression).
8721 Default is @var{none}.
8724 @subsection Examples
8728 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
8729 using @command{ffmpeg}:
8731 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
8735 @section perspective
8737 Correct perspective of video not recorded perpendicular to the screen.
8739 A description of the accepted parameters follows.
8750 Set coordinates expression for top left, top right, bottom left and bottom right corners.
8751 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
8752 If the @code{sense} option is set to @code{source}, then the specified points will be sent
8753 to the corners of the destination. If the @code{sense} option is set to @code{destination},
8754 then the corners of the source will be sent to the specified coordinates.
8756 The expressions can use the following variables:
8761 the width and height of video frame.
8765 Set interpolation for perspective correction.
8767 It accepts the following values:
8773 Default value is @samp{linear}.
8776 Set interpretation of coordinate options.
8778 It accepts the following values:
8782 Send point in the source specified by the given coordinates to
8783 the corners of the destination.
8785 @item 1, destination
8787 Send the corners of the source to the point in the destination specified
8788 by the given coordinates.
8790 Default value is @samp{source}.
8796 Delay interlaced video by one field time so that the field order changes.
8798 The intended use is to fix PAL movies that have been captured with the
8799 opposite field order to the film-to-video transfer.
8801 A description of the accepted parameters follows.
8807 It accepts the following values:
8810 Capture field order top-first, transfer bottom-first.
8811 Filter will delay the bottom field.
8814 Capture field order bottom-first, transfer top-first.
8815 Filter will delay the top field.
8818 Capture and transfer with the same field order. This mode only exists
8819 for the documentation of the other options to refer to, but if you
8820 actually select it, the filter will faithfully do nothing.
8823 Capture field order determined automatically by field flags, transfer
8825 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
8826 basis using field flags. If no field information is available,
8827 then this works just like @samp{u}.
8830 Capture unknown or varying, transfer opposite.
8831 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
8832 analyzing the images and selecting the alternative that produces best
8833 match between the fields.
8836 Capture top-first, transfer unknown or varying.
8837 Filter selects among @samp{t} and @samp{p} using image analysis.
8840 Capture bottom-first, transfer unknown or varying.
8841 Filter selects among @samp{b} and @samp{p} using image analysis.
8844 Capture determined by field flags, transfer unknown or varying.
8845 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
8846 image analysis. If no field information is available, then this works just
8847 like @samp{U}. This is the default mode.
8850 Both capture and transfer unknown or varying.
8851 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
8855 @section pixdesctest
8857 Pixel format descriptor test filter, mainly useful for internal
8858 testing. The output video should be equal to the input video.
8862 format=monow, pixdesctest
8865 can be used to test the monowhite pixel format descriptor definition.
8869 Enable the specified chain of postprocessing subfilters using libpostproc. This
8870 library should be automatically selected with a GPL build (@code{--enable-gpl}).
8871 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
8872 Each subfilter and some options have a short and a long name that can be used
8873 interchangeably, i.e. dr/dering are the same.
8875 The filters accept the following options:
8879 Set postprocessing subfilters string.
8882 All subfilters share common options to determine their scope:
8886 Honor the quality commands for this subfilter.
8889 Do chrominance filtering, too (default).
8892 Do luminance filtering only (no chrominance).
8895 Do chrominance filtering only (no luminance).
8898 These options can be appended after the subfilter name, separated by a '|'.
8900 Available subfilters are:
8903 @item hb/hdeblock[|difference[|flatness]]
8904 Horizontal deblocking filter
8907 Difference factor where higher values mean more deblocking (default: @code{32}).
8909 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8912 @item vb/vdeblock[|difference[|flatness]]
8913 Vertical deblocking filter
8916 Difference factor where higher values mean more deblocking (default: @code{32}).
8918 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8921 @item ha/hadeblock[|difference[|flatness]]
8922 Accurate horizontal deblocking filter
8925 Difference factor where higher values mean more deblocking (default: @code{32}).
8927 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8930 @item va/vadeblock[|difference[|flatness]]
8931 Accurate vertical deblocking filter
8934 Difference factor where higher values mean more deblocking (default: @code{32}).
8936 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8940 The horizontal and vertical deblocking filters share the difference and
8941 flatness values so you cannot set different horizontal and vertical
8946 Experimental horizontal deblocking filter
8949 Experimental vertical deblocking filter
8954 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
8957 larger -> stronger filtering
8959 larger -> stronger filtering
8961 larger -> stronger filtering
8964 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
8967 Stretch luminance to @code{0-255}.
8970 @item lb/linblenddeint
8971 Linear blend deinterlacing filter that deinterlaces the given block by
8972 filtering all lines with a @code{(1 2 1)} filter.
8974 @item li/linipoldeint
8975 Linear interpolating deinterlacing filter that deinterlaces the given block by
8976 linearly interpolating every second line.
8978 @item ci/cubicipoldeint
8979 Cubic interpolating deinterlacing filter deinterlaces the given block by
8980 cubically interpolating every second line.
8982 @item md/mediandeint
8983 Median deinterlacing filter that deinterlaces the given block by applying a
8984 median filter to every second line.
8986 @item fd/ffmpegdeint
8987 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
8988 second line with a @code{(-1 4 2 4 -1)} filter.
8991 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
8992 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
8994 @item fq/forceQuant[|quantizer]
8995 Overrides the quantizer table from the input with the constant quantizer you
9003 Default pp filter combination (@code{hb|a,vb|a,dr|a})
9006 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
9009 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
9012 @subsection Examples
9016 Apply horizontal and vertical deblocking, deringing and automatic
9017 brightness/contrast:
9023 Apply default filters without brightness/contrast correction:
9029 Apply default filters and temporal denoiser:
9031 pp=default/tmpnoise|1|2|3
9035 Apply deblocking on luminance only, and switch vertical deblocking on or off
9036 automatically depending on available CPU time:
9043 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
9044 similar to spp = 6 with 7 point DCT, where only the center sample is
9047 The filter accepts the following options:
9051 Force a constant quantization parameter. It accepts an integer in range
9052 0 to 63. If not set, the filter will use the QP from the video stream
9056 Set thresholding mode. Available modes are:
9060 Set hard thresholding.
9062 Set soft thresholding (better de-ringing effect, but likely blurrier).
9064 Set medium thresholding (good results, default).
9070 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
9071 Ratio) between two input videos.
9073 This filter takes in input two input videos, the first input is
9074 considered the "main" source and is passed unchanged to the
9075 output. The second input is used as a "reference" video for computing
9078 Both video inputs must have the same resolution and pixel format for
9079 this filter to work correctly. Also it assumes that both inputs
9080 have the same number of frames, which are compared one by one.
9082 The obtained average PSNR is printed through the logging system.
9084 The filter stores the accumulated MSE (mean squared error) of each
9085 frame, and at the end of the processing it is averaged across all frames
9086 equally, and the following formula is applied to obtain the PSNR:
9089 PSNR = 10*log10(MAX^2/MSE)
9092 Where MAX is the average of the maximum values of each component of the
9095 The description of the accepted parameters follows.
9099 If specified the filter will use the named file to save the PSNR of
9100 each individual frame. When filename equals "-" the data is sent to
9104 The file printed if @var{stats_file} is selected, contains a sequence of
9105 key/value pairs of the form @var{key}:@var{value} for each compared
9108 A description of each shown parameter follows:
9112 sequential number of the input frame, starting from 1
9115 Mean Square Error pixel-by-pixel average difference of the compared
9116 frames, averaged over all the image components.
9118 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
9119 Mean Square Error pixel-by-pixel average difference of the compared
9120 frames for the component specified by the suffix.
9122 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
9123 Peak Signal to Noise ratio of the compared frames for the component
9124 specified by the suffix.
9129 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
9130 [main][ref] psnr="stats_file=stats.log" [out]
9133 On this example the input file being processed is compared with the
9134 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
9135 is stored in @file{stats.log}.
9140 Pulldown reversal (inverse telecine) filter, capable of handling mixed
9141 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
9144 The pullup filter is designed to take advantage of future context in making
9145 its decisions. This filter is stateless in the sense that it does not lock
9146 onto a pattern to follow, but it instead looks forward to the following
9147 fields in order to identify matches and rebuild progressive frames.
9149 To produce content with an even framerate, insert the fps filter after
9150 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
9151 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
9153 The filter accepts the following options:
9160 These options set the amount of "junk" to ignore at the left, right, top, and
9161 bottom of the image, respectively. Left and right are in units of 8 pixels,
9162 while top and bottom are in units of 2 lines.
9163 The default is 8 pixels on each side.
9166 Set the strict breaks. Setting this option to 1 will reduce the chances of
9167 filter generating an occasional mismatched frame, but it may also cause an
9168 excessive number of frames to be dropped during high motion sequences.
9169 Conversely, setting it to -1 will make filter match fields more easily.
9170 This may help processing of video where there is slight blurring between
9171 the fields, but may also cause there to be interlaced frames in the output.
9172 Default value is @code{0}.
9175 Set the metric plane to use. It accepts the following values:
9181 Use chroma blue plane.
9184 Use chroma red plane.
9187 This option may be set to use chroma plane instead of the default luma plane
9188 for doing filter's computations. This may improve accuracy on very clean
9189 source material, but more likely will decrease accuracy, especially if there
9190 is chroma noise (rainbow effect) or any grayscale video.
9191 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
9192 load and make pullup usable in realtime on slow machines.
9195 For best results (without duplicated frames in the output file) it is
9196 necessary to change the output frame rate. For example, to inverse
9197 telecine NTSC input:
9199 ffmpeg -i input -vf pullup -r 24000/1001 ...
9204 Change video quantization parameters (QP).
9206 The filter accepts the following option:
9210 Set expression for quantization parameter.
9213 The expression is evaluated through the eval API and can contain, among others,
9214 the following constants:
9218 1 if index is not 129, 0 otherwise.
9221 Sequentional index starting from -129 to 128.
9224 @subsection Examples
9236 Flush video frames from internal cache of frames into a random order.
9237 No frame is discarded.
9238 Inspired by @ref{frei0r} nervous filter.
9242 Set size in number of frames of internal cache, in range from @code{2} to
9243 @code{512}. Default is @code{30}.
9246 Set seed for random number generator, must be an integer included between
9247 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
9248 less than @code{0}, the filter will try to use a good random seed on a
9252 @section removegrain
9254 The removegrain filter is a spatial denoiser for progressive video.
9258 Set mode for the first plane.
9261 Set mode for the second plane.
9264 Set mode for the third plane.
9267 Set mode for the fourth plane.
9270 Range of mode is from 0 to 24. Description of each mode follows:
9274 Leave input plane unchanged. Default.
9277 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
9280 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
9283 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
9286 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
9287 This is equivalent to a median filter.
9290 Line-sensitive clipping giving the minimal change.
9293 Line-sensitive clipping, intermediate.
9296 Line-sensitive clipping, intermediate.
9299 Line-sensitive clipping, intermediate.
9302 Line-sensitive clipping on a line where the neighbours pixels are the closest.
9305 Replaces the target pixel with the closest neighbour.
9308 [1 2 1] horizontal and vertical kernel blur.
9314 Bob mode, interpolates top field from the line where the neighbours
9315 pixels are the closest.
9318 Bob mode, interpolates bottom field from the line where the neighbours
9319 pixels are the closest.
9322 Bob mode, interpolates top field. Same as 13 but with a more complicated
9323 interpolation formula.
9326 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
9327 interpolation formula.
9330 Clips the pixel with the minimum and maximum of respectively the maximum and
9331 minimum of each pair of opposite neighbour pixels.
9334 Line-sensitive clipping using opposite neighbours whose greatest distance from
9335 the current pixel is minimal.
9338 Replaces the pixel with the average of its 8 neighbours.
9341 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
9344 Clips pixels using the averages of opposite neighbour.
9347 Same as mode 21 but simpler and faster.
9350 Small edge and halo removal, but reputed useless.
9358 Suppress a TV station logo, using an image file to determine which
9359 pixels comprise the logo. It works by filling in the pixels that
9360 comprise the logo with neighboring pixels.
9362 The filter accepts the following options:
9366 Set the filter bitmap file, which can be any image format supported by
9367 libavformat. The width and height of the image file must match those of the
9368 video stream being processed.
9371 Pixels in the provided bitmap image with a value of zero are not
9372 considered part of the logo, non-zero pixels are considered part of
9373 the logo. If you use white (255) for the logo and black (0) for the
9374 rest, you will be safe. For making the filter bitmap, it is
9375 recommended to take a screen capture of a black frame with the logo
9376 visible, and then using a threshold filter followed by the erode
9377 filter once or twice.
9379 If needed, little splotches can be fixed manually. Remember that if
9380 logo pixels are not covered, the filter quality will be much
9381 reduced. Marking too many pixels as part of the logo does not hurt as
9382 much, but it will increase the amount of blurring needed to cover over
9383 the image and will destroy more information than necessary, and extra
9384 pixels will slow things down on a large logo.
9386 @section repeatfields
9388 This filter uses the repeat_field flag from the Video ES headers and hard repeats
9389 fields based on its value.
9391 @section reverse, areverse
9395 Warning: This filter requires memory to buffer the entire clip, so trimming
9398 @subsection Examples
9402 Take the first 5 seconds of a clip, and reverse it.
9410 Rotate video by an arbitrary angle expressed in radians.
9412 The filter accepts the following options:
9414 A description of the optional parameters follows.
9417 Set an expression for the angle by which to rotate the input video
9418 clockwise, expressed as a number of radians. A negative value will
9419 result in a counter-clockwise rotation. By default it is set to "0".
9421 This expression is evaluated for each frame.
9424 Set the output width expression, default value is "iw".
9425 This expression is evaluated just once during configuration.
9428 Set the output height expression, default value is "ih".
9429 This expression is evaluated just once during configuration.
9432 Enable bilinear interpolation if set to 1, a value of 0 disables
9433 it. Default value is 1.
9436 Set the color used to fill the output area not covered by the rotated
9437 image. For the general syntax of this option, check the "Color" section in the
9438 ffmpeg-utils manual. If the special value "none" is selected then no
9439 background is printed (useful for example if the background is never shown).
9441 Default value is "black".
9444 The expressions for the angle and the output size can contain the
9445 following constants and functions:
9449 sequential number of the input frame, starting from 0. It is always NAN
9450 before the first frame is filtered.
9453 time in seconds of the input frame, it is set to 0 when the filter is
9454 configured. It is always NAN before the first frame is filtered.
9458 horizontal and vertical chroma subsample values. For example for the
9459 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9463 the input video width and height
9467 the output width and height, that is the size of the padded area as
9468 specified by the @var{width} and @var{height} expressions
9472 the minimal width/height required for completely containing the input
9473 video rotated by @var{a} radians.
9475 These are only available when computing the @option{out_w} and
9476 @option{out_h} expressions.
9479 @subsection Examples
9483 Rotate the input by PI/6 radians clockwise:
9489 Rotate the input by PI/6 radians counter-clockwise:
9495 Rotate the input by 45 degrees clockwise:
9501 Apply a constant rotation with period T, starting from an angle of PI/3:
9503 rotate=PI/3+2*PI*t/T
9507 Make the input video rotation oscillating with a period of T
9508 seconds and an amplitude of A radians:
9510 rotate=A*sin(2*PI/T*t)
9514 Rotate the video, output size is chosen so that the whole rotating
9515 input video is always completely contained in the output:
9517 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
9521 Rotate the video, reduce the output size so that no background is ever
9524 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
9528 @subsection Commands
9530 The filter supports the following commands:
9534 Set the angle expression.
9535 The command accepts the same syntax of the corresponding option.
9537 If the specified expression is not valid, it is kept at its current
9543 Apply Shape Adaptive Blur.
9545 The filter accepts the following options:
9548 @item luma_radius, lr
9549 Set luma blur filter strength, must be a value in range 0.1-4.0, default
9550 value is 1.0. A greater value will result in a more blurred image, and
9551 in slower processing.
9553 @item luma_pre_filter_radius, lpfr
9554 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
9557 @item luma_strength, ls
9558 Set luma maximum difference between pixels to still be considered, must
9559 be a value in the 0.1-100.0 range, default value is 1.0.
9561 @item chroma_radius, cr
9562 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
9563 greater value will result in a more blurred image, and in slower
9566 @item chroma_pre_filter_radius, cpfr
9567 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
9569 @item chroma_strength, cs
9570 Set chroma maximum difference between pixels to still be considered,
9571 must be a value in the 0.1-100.0 range.
9574 Each chroma option value, if not explicitly specified, is set to the
9575 corresponding luma option value.
9580 Scale (resize) the input video, using the libswscale library.
9582 The scale filter forces the output display aspect ratio to be the same
9583 of the input, by changing the output sample aspect ratio.
9585 If the input image format is different from the format requested by
9586 the next filter, the scale filter will convert the input to the
9590 The filter accepts the following options, or any of the options
9591 supported by the libswscale scaler.
9593 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
9594 the complete list of scaler options.
9599 Set the output video dimension expression. Default value is the input
9602 If the value is 0, the input width is used for the output.
9604 If one of the values is -1, the scale filter will use a value that
9605 maintains the aspect ratio of the input image, calculated from the
9606 other specified dimension. If both of them are -1, the input size is
9609 If one of the values is -n with n > 1, the scale filter will also use a value
9610 that maintains the aspect ratio of the input image, calculated from the other
9611 specified dimension. After that it will, however, make sure that the calculated
9612 dimension is divisible by n and adjust the value if necessary.
9614 See below for the list of accepted constants for use in the dimension
9618 Set the interlacing mode. It accepts the following values:
9622 Force interlaced aware scaling.
9625 Do not apply interlaced scaling.
9628 Select interlaced aware scaling depending on whether the source frames
9629 are flagged as interlaced or not.
9632 Default value is @samp{0}.
9635 Set libswscale scaling flags. See
9636 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
9637 complete list of values. If not explicitly specified the filter applies
9641 Set the video size. For the syntax of this option, check the
9642 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9644 @item in_color_matrix
9645 @item out_color_matrix
9646 Set in/output YCbCr color space type.
9648 This allows the autodetected value to be overridden as well as allows forcing
9649 a specific value used for the output and encoder.
9651 If not specified, the color space type depends on the pixel format.
9657 Choose automatically.
9660 Format conforming to International Telecommunication Union (ITU)
9661 Recommendation BT.709.
9664 Set color space conforming to the United States Federal Communications
9665 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
9668 Set color space conforming to:
9672 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
9675 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
9678 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
9683 Set color space conforming to SMPTE ST 240:1999.
9688 Set in/output YCbCr sample range.
9690 This allows the autodetected value to be overridden as well as allows forcing
9691 a specific value used for the output and encoder. If not specified, the
9692 range depends on the pixel format. Possible values:
9696 Choose automatically.
9699 Set full range (0-255 in case of 8-bit luma).
9702 Set "MPEG" range (16-235 in case of 8-bit luma).
9705 @item force_original_aspect_ratio
9706 Enable decreasing or increasing output video width or height if necessary to
9707 keep the original aspect ratio. Possible values:
9711 Scale the video as specified and disable this feature.
9714 The output video dimensions will automatically be decreased if needed.
9717 The output video dimensions will automatically be increased if needed.
9721 One useful instance of this option is that when you know a specific device's
9722 maximum allowed resolution, you can use this to limit the output video to
9723 that, while retaining the aspect ratio. For example, device A allows
9724 1280x720 playback, and your video is 1920x800. Using this option (set it to
9725 decrease) and specifying 1280x720 to the command line makes the output
9728 Please note that this is a different thing than specifying -1 for @option{w}
9729 or @option{h}, you still need to specify the output resolution for this option
9734 The values of the @option{w} and @option{h} options are expressions
9735 containing the following constants:
9740 The input width and height
9744 These are the same as @var{in_w} and @var{in_h}.
9748 The output (scaled) width and height
9752 These are the same as @var{out_w} and @var{out_h}
9755 The same as @var{iw} / @var{ih}
9758 input sample aspect ratio
9761 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
9765 horizontal and vertical input chroma subsample values. For example for the
9766 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9770 horizontal and vertical output chroma subsample values. For example for the
9771 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9774 @subsection Examples
9778 Scale the input video to a size of 200x100
9783 This is equivalent to:
9794 Specify a size abbreviation for the output size:
9799 which can also be written as:
9805 Scale the input to 2x:
9811 The above is the same as:
9817 Scale the input to 2x with forced interlaced scaling:
9819 scale=2*iw:2*ih:interl=1
9823 Scale the input to half size:
9829 Increase the width, and set the height to the same size:
9842 Increase the height, and set the width to 3/2 of the height:
9844 scale=w=3/2*oh:h=3/5*ih
9848 Increase the size, making the size a multiple of the chroma
9851 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
9855 Increase the width to a maximum of 500 pixels,
9856 keeping the same aspect ratio as the input:
9858 scale=w='min(500\, iw*3/2):h=-1'
9862 @subsection Commands
9864 This filter supports the following commands:
9868 Set the output video dimension expression.
9869 The command accepts the same syntax of the corresponding option.
9871 If the specified expression is not valid, it is kept at its current
9877 Scale (resize) the input video, based on a reference video.
9879 See the scale filter for available options, scale2ref supports the same but
9880 uses the reference video instead of the main input as basis.
9882 @subsection Examples
9886 Scale a subtitle stream to match the main video in size before overlaying
9888 'scale2ref[b][a];[a][b]overlay'
9892 @section separatefields
9894 The @code{separatefields} takes a frame-based video input and splits
9895 each frame into its components fields, producing a new half height clip
9896 with twice the frame rate and twice the frame count.
9898 This filter use field-dominance information in frame to decide which
9899 of each pair of fields to place first in the output.
9900 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
9902 @section setdar, setsar
9904 The @code{setdar} filter sets the Display Aspect Ratio for the filter
9907 This is done by changing the specified Sample (aka Pixel) Aspect
9908 Ratio, according to the following equation:
9910 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
9913 Keep in mind that the @code{setdar} filter does not modify the pixel
9914 dimensions of the video frame. Also, the display aspect ratio set by
9915 this filter may be changed by later filters in the filterchain,
9916 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
9919 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
9920 the filter output video.
9922 Note that as a consequence of the application of this filter, the
9923 output display aspect ratio will change according to the equation
9926 Keep in mind that the sample aspect ratio set by the @code{setsar}
9927 filter may be changed by later filters in the filterchain, e.g. if
9928 another "setsar" or a "setdar" filter is applied.
9930 It accepts the following parameters:
9933 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
9934 Set the aspect ratio used by the filter.
9936 The parameter can be a floating point number string, an expression, or
9937 a string of the form @var{num}:@var{den}, where @var{num} and
9938 @var{den} are the numerator and denominator of the aspect ratio. If
9939 the parameter is not specified, it is assumed the value "0".
9940 In case the form "@var{num}:@var{den}" is used, the @code{:} character
9944 Set the maximum integer value to use for expressing numerator and
9945 denominator when reducing the expressed aspect ratio to a rational.
9946 Default value is @code{100}.
9950 The parameter @var{sar} is an expression containing
9951 the following constants:
9955 These are approximated values for the mathematical constants e
9956 (Euler's number), pi (Greek pi), and phi (the golden ratio).
9959 The input width and height.
9962 These are the same as @var{w} / @var{h}.
9965 The input sample aspect ratio.
9968 The input display aspect ratio. It is the same as
9969 (@var{w} / @var{h}) * @var{sar}.
9972 Horizontal and vertical chroma subsample values. For example, for the
9973 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9976 @subsection Examples
9981 To change the display aspect ratio to 16:9, specify one of the following:
9989 To change the sample aspect ratio to 10:11, specify:
9995 To set a display aspect ratio of 16:9, and specify a maximum integer value of
9996 1000 in the aspect ratio reduction, use the command:
9998 setdar=ratio=16/9:max=1000
10006 Force field for the output video frame.
10008 The @code{setfield} filter marks the interlace type field for the
10009 output frames. It does not change the input frame, but only sets the
10010 corresponding property, which affects how the frame is treated by
10011 following filters (e.g. @code{fieldorder} or @code{yadif}).
10013 The filter accepts the following options:
10018 Available values are:
10022 Keep the same field property.
10025 Mark the frame as bottom-field-first.
10028 Mark the frame as top-field-first.
10031 Mark the frame as progressive.
10037 Show a line containing various information for each input video frame.
10038 The input video is not modified.
10040 The shown line contains a sequence of key/value pairs of the form
10041 @var{key}:@var{value}.
10043 The following values are shown in the output:
10047 The (sequential) number of the input frame, starting from 0.
10050 The Presentation TimeStamp of the input frame, expressed as a number of
10051 time base units. The time base unit depends on the filter input pad.
10054 The Presentation TimeStamp of the input frame, expressed as a number of
10058 The position of the frame in the input stream, or -1 if this information is
10059 unavailable and/or meaningless (for example in case of synthetic video).
10062 The pixel format name.
10065 The sample aspect ratio of the input frame, expressed in the form
10066 @var{num}/@var{den}.
10069 The size of the input frame. For the syntax of this option, check the
10070 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10073 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
10074 for bottom field first).
10077 This is 1 if the frame is a key frame, 0 otherwise.
10080 The picture type of the input frame ("I" for an I-frame, "P" for a
10081 P-frame, "B" for a B-frame, or "?" for an unknown type).
10082 Also refer to the documentation of the @code{AVPictureType} enum and of
10083 the @code{av_get_picture_type_char} function defined in
10084 @file{libavutil/avutil.h}.
10087 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
10089 @item plane_checksum
10090 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
10091 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
10094 @section showpalette
10096 Displays the 256 colors palette of each frame. This filter is only relevant for
10097 @var{pal8} pixel format frames.
10099 It accepts the following option:
10103 Set the size of the box used to represent one palette color entry. Default is
10104 @code{30} (for a @code{30x30} pixel box).
10107 @section shuffleframes
10109 Reorder and/or duplicate video frames.
10111 It accepts the following parameters:
10115 Set the destination indexes of input frames.
10116 This is space or '|' separated list of indexes that maps input frames to output
10117 frames. Number of indexes also sets maximal value that each index may have.
10120 The first frame has the index 0. The default is to keep the input unchanged.
10122 Swap second and third frame of every three frames of the input:
10124 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
10127 @section shuffleplanes
10129 Reorder and/or duplicate video planes.
10131 It accepts the following parameters:
10136 The index of the input plane to be used as the first output plane.
10139 The index of the input plane to be used as the second output plane.
10142 The index of the input plane to be used as the third output plane.
10145 The index of the input plane to be used as the fourth output plane.
10149 The first plane has the index 0. The default is to keep the input unchanged.
10151 Swap the second and third planes of the input:
10153 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
10156 @anchor{signalstats}
10157 @section signalstats
10158 Evaluate various visual metrics that assist in determining issues associated
10159 with the digitization of analog video media.
10161 By default the filter will log these metadata values:
10165 Display the minimal Y value contained within the input frame. Expressed in
10169 Display the Y value at the 10% percentile within the input frame. Expressed in
10173 Display the average Y value within the input frame. Expressed in range of
10177 Display the Y value at the 90% percentile within the input frame. Expressed in
10181 Display the maximum Y value contained within the input frame. Expressed in
10185 Display the minimal U value contained within the input frame. Expressed in
10189 Display the U value at the 10% percentile within the input frame. Expressed in
10193 Display the average U value within the input frame. Expressed in range of
10197 Display the U value at the 90% percentile within the input frame. Expressed in
10201 Display the maximum U value contained within the input frame. Expressed in
10205 Display the minimal V value contained within the input frame. Expressed in
10209 Display the V value at the 10% percentile within the input frame. Expressed in
10213 Display the average V value within the input frame. Expressed in range of
10217 Display the V value at the 90% percentile within the input frame. Expressed in
10221 Display the maximum V value contained within the input frame. Expressed in
10225 Display the minimal saturation value contained within the input frame.
10226 Expressed in range of [0-~181.02].
10229 Display the saturation value at the 10% percentile within the input frame.
10230 Expressed in range of [0-~181.02].
10233 Display the average saturation value within the input frame. Expressed in range
10237 Display the saturation value at the 90% percentile within the input frame.
10238 Expressed in range of [0-~181.02].
10241 Display the maximum saturation value contained within the input frame.
10242 Expressed in range of [0-~181.02].
10245 Display the median value for hue within the input frame. Expressed in range of
10249 Display the average value for hue within the input frame. Expressed in range of
10253 Display the average of sample value difference between all values of the Y
10254 plane in the current frame and corresponding values of the previous input frame.
10255 Expressed in range of [0-255].
10258 Display the average of sample value difference between all values of the U
10259 plane in the current frame and corresponding values of the previous input frame.
10260 Expressed in range of [0-255].
10263 Display the average of sample value difference between all values of the V
10264 plane in the current frame and corresponding values of the previous input frame.
10265 Expressed in range of [0-255].
10268 The filter accepts the following options:
10274 @option{stat} specify an additional form of image analysis.
10275 @option{out} output video with the specified type of pixel highlighted.
10277 Both options accept the following values:
10281 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
10282 unlike the neighboring pixels of the same field. Examples of temporal outliers
10283 include the results of video dropouts, head clogs, or tape tracking issues.
10286 Identify @var{vertical line repetition}. Vertical line repetition includes
10287 similar rows of pixels within a frame. In born-digital video vertical line
10288 repetition is common, but this pattern is uncommon in video digitized from an
10289 analog source. When it occurs in video that results from the digitization of an
10290 analog source it can indicate concealment from a dropout compensator.
10293 Identify pixels that fall outside of legal broadcast range.
10297 Set the highlight color for the @option{out} option. The default color is
10301 @subsection Examples
10305 Output data of various video metrics:
10307 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
10311 Output specific data about the minimum and maximum values of the Y plane per frame:
10313 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
10317 Playback video while highlighting pixels that are outside of broadcast range in red.
10319 ffplay example.mov -vf signalstats="out=brng:color=red"
10323 Playback video with signalstats metadata drawn over the frame.
10325 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
10328 The contents of signalstat_drawtext.txt used in the command are:
10331 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
10332 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
10333 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
10334 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
10342 Blur the input video without impacting the outlines.
10344 It accepts the following options:
10347 @item luma_radius, lr
10348 Set the luma radius. The option value must be a float number in
10349 the range [0.1,5.0] that specifies the variance of the gaussian filter
10350 used to blur the image (slower if larger). Default value is 1.0.
10352 @item luma_strength, ls
10353 Set the luma strength. The option value must be a float number
10354 in the range [-1.0,1.0] that configures the blurring. A value included
10355 in [0.0,1.0] will blur the image whereas a value included in
10356 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10358 @item luma_threshold, lt
10359 Set the luma threshold used as a coefficient to determine
10360 whether a pixel should be blurred or not. The option value must be an
10361 integer in the range [-30,30]. A value of 0 will filter all the image,
10362 a value included in [0,30] will filter flat areas and a value included
10363 in [-30,0] will filter edges. Default value is 0.
10365 @item chroma_radius, cr
10366 Set the chroma radius. The option value must be a float number in
10367 the range [0.1,5.0] that specifies the variance of the gaussian filter
10368 used to blur the image (slower if larger). Default value is 1.0.
10370 @item chroma_strength, cs
10371 Set the chroma strength. The option value must be a float number
10372 in the range [-1.0,1.0] that configures the blurring. A value included
10373 in [0.0,1.0] will blur the image whereas a value included in
10374 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10376 @item chroma_threshold, ct
10377 Set the chroma threshold used as a coefficient to determine
10378 whether a pixel should be blurred or not. The option value must be an
10379 integer in the range [-30,30]. A value of 0 will filter all the image,
10380 a value included in [0,30] will filter flat areas and a value included
10381 in [-30,0] will filter edges. Default value is 0.
10384 If a chroma option is not explicitly set, the corresponding luma value
10389 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
10391 This filter takes in input two input videos, the first input is
10392 considered the "main" source and is passed unchanged to the
10393 output. The second input is used as a "reference" video for computing
10396 Both video inputs must have the same resolution and pixel format for
10397 this filter to work correctly. Also it assumes that both inputs
10398 have the same number of frames, which are compared one by one.
10400 The filter stores the calculated SSIM of each frame.
10402 The description of the accepted parameters follows.
10405 @item stats_file, f
10406 If specified the filter will use the named file to save the SSIM of
10407 each individual frame. When filename equals "-" the data is sent to
10411 The file printed if @var{stats_file} is selected, contains a sequence of
10412 key/value pairs of the form @var{key}:@var{value} for each compared
10415 A description of each shown parameter follows:
10419 sequential number of the input frame, starting from 1
10421 @item Y, U, V, R, G, B
10422 SSIM of the compared frames for the component specified by the suffix.
10425 SSIM of the compared frames for the whole frame.
10428 Same as above but in dB representation.
10433 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
10434 [main][ref] ssim="stats_file=stats.log" [out]
10437 On this example the input file being processed is compared with the
10438 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
10439 is stored in @file{stats.log}.
10441 Another example with both psnr and ssim at same time:
10443 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
10448 Convert between different stereoscopic image formats.
10450 The filters accept the following options:
10454 Set stereoscopic image format of input.
10456 Available values for input image formats are:
10459 side by side parallel (left eye left, right eye right)
10462 side by side crosseye (right eye left, left eye right)
10465 side by side parallel with half width resolution
10466 (left eye left, right eye right)
10469 side by side crosseye with half width resolution
10470 (right eye left, left eye right)
10473 above-below (left eye above, right eye below)
10476 above-below (right eye above, left eye below)
10479 above-below with half height resolution
10480 (left eye above, right eye below)
10483 above-below with half height resolution
10484 (right eye above, left eye below)
10487 alternating frames (left eye first, right eye second)
10490 alternating frames (right eye first, left eye second)
10493 interleaved rows (left eye has top row, right eye starts on next row)
10496 interleaved rows (right eye has top row, left eye starts on next row)
10498 Default value is @samp{sbsl}.
10502 Set stereoscopic image format of output.
10504 Available values for output image formats are all the input formats as well as:
10507 anaglyph red/blue gray
10508 (red filter on left eye, blue filter on right eye)
10511 anaglyph red/green gray
10512 (red filter on left eye, green filter on right eye)
10515 anaglyph red/cyan gray
10516 (red filter on left eye, cyan filter on right eye)
10519 anaglyph red/cyan half colored
10520 (red filter on left eye, cyan filter on right eye)
10523 anaglyph red/cyan color
10524 (red filter on left eye, cyan filter on right eye)
10527 anaglyph red/cyan color optimized with the least squares projection of dubois
10528 (red filter on left eye, cyan filter on right eye)
10531 anaglyph green/magenta gray
10532 (green filter on left eye, magenta filter on right eye)
10535 anaglyph green/magenta half colored
10536 (green filter on left eye, magenta filter on right eye)
10539 anaglyph green/magenta colored
10540 (green filter on left eye, magenta filter on right eye)
10543 anaglyph green/magenta color optimized with the least squares projection of dubois
10544 (green filter on left eye, magenta filter on right eye)
10547 anaglyph yellow/blue gray
10548 (yellow filter on left eye, blue filter on right eye)
10551 anaglyph yellow/blue half colored
10552 (yellow filter on left eye, blue filter on right eye)
10555 anaglyph yellow/blue colored
10556 (yellow filter on left eye, blue filter on right eye)
10559 anaglyph yellow/blue color optimized with the least squares projection of dubois
10560 (yellow filter on left eye, blue filter on right eye)
10563 mono output (left eye only)
10566 mono output (right eye only)
10569 checkerboard, left eye first
10572 checkerboard, right eye first
10575 interleaved columns, left eye first
10578 interleaved columns, right eye first
10581 Default value is @samp{arcd}.
10584 @subsection Examples
10588 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
10594 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
10603 Apply a simple postprocessing filter that compresses and decompresses the image
10604 at several (or - in the case of @option{quality} level @code{6} - all) shifts
10605 and average the results.
10607 The filter accepts the following options:
10611 Set quality. This option defines the number of levels for averaging. It accepts
10612 an integer in the range 0-6. If set to @code{0}, the filter will have no
10613 effect. A value of @code{6} means the higher quality. For each increment of
10614 that value the speed drops by a factor of approximately 2. Default value is
10618 Force a constant quantization parameter. If not set, the filter will use the QP
10619 from the video stream (if available).
10622 Set thresholding mode. Available modes are:
10626 Set hard thresholding (default).
10628 Set soft thresholding (better de-ringing effect, but likely blurrier).
10631 @item use_bframe_qp
10632 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
10633 option may cause flicker since the B-Frames have often larger QP. Default is
10634 @code{0} (not enabled).
10640 Draw subtitles on top of input video using the libass library.
10642 To enable compilation of this filter you need to configure FFmpeg with
10643 @code{--enable-libass}. This filter also requires a build with libavcodec and
10644 libavformat to convert the passed subtitles file to ASS (Advanced Substation
10645 Alpha) subtitles format.
10647 The filter accepts the following options:
10651 Set the filename of the subtitle file to read. It must be specified.
10653 @item original_size
10654 Specify the size of the original video, the video for which the ASS file
10655 was composed. For the syntax of this option, check the
10656 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10657 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
10658 correctly scale the fonts if the aspect ratio has been changed.
10661 Set a directory path containing fonts that can be used by the filter.
10662 These fonts will be used in addition to whatever the font provider uses.
10665 Set subtitles input character encoding. @code{subtitles} filter only. Only
10666 useful if not UTF-8.
10668 @item stream_index, si
10669 Set subtitles stream index. @code{subtitles} filter only.
10672 Override default style or script info parameters of the subtitles. It accepts a
10673 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
10676 If the first key is not specified, it is assumed that the first value
10677 specifies the @option{filename}.
10679 For example, to render the file @file{sub.srt} on top of the input
10680 video, use the command:
10685 which is equivalent to:
10687 subtitles=filename=sub.srt
10690 To render the default subtitles stream from file @file{video.mkv}, use:
10692 subtitles=video.mkv
10695 To render the second subtitles stream from that file, use:
10697 subtitles=video.mkv:si=1
10700 To make the subtitles stream from @file{sub.srt} appear in transparent green
10701 @code{DejaVu Serif}, use:
10703 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
10706 @section super2xsai
10708 Scale the input by 2x and smooth using the Super2xSaI (Scale and
10709 Interpolate) pixel art scaling algorithm.
10711 Useful for enlarging pixel art images without reducing sharpness.
10718 Apply telecine process to the video.
10720 This filter accepts the following options:
10729 The default value is @code{top}.
10733 A string of numbers representing the pulldown pattern you wish to apply.
10734 The default value is @code{23}.
10738 Some typical patterns:
10743 24p: 2332 (preferred)
10750 24p: 222222222223 ("Euro pulldown")
10756 Select the most representative frame in a given sequence of consecutive frames.
10758 The filter accepts the following options:
10762 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
10763 will pick one of them, and then handle the next batch of @var{n} frames until
10764 the end. Default is @code{100}.
10767 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
10768 value will result in a higher memory usage, so a high value is not recommended.
10770 @subsection Examples
10774 Extract one picture each 50 frames:
10780 Complete example of a thumbnail creation with @command{ffmpeg}:
10782 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
10788 Tile several successive frames together.
10790 The filter accepts the following options:
10795 Set the grid size (i.e. the number of lines and columns). For the syntax of
10796 this option, check the
10797 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10800 Set the maximum number of frames to render in the given area. It must be less
10801 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
10802 the area will be used.
10805 Set the outer border margin in pixels.
10808 Set the inner border thickness (i.e. the number of pixels between frames). For
10809 more advanced padding options (such as having different values for the edges),
10810 refer to the pad video filter.
10813 Specify the color of the unused area. For the syntax of this option, check the
10814 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
10818 @subsection Examples
10822 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
10824 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
10826 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
10827 duplicating each output frame to accommodate the originally detected frame
10831 Display @code{5} pictures in an area of @code{3x2} frames,
10832 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
10833 mixed flat and named options:
10835 tile=3x2:nb_frames=5:padding=7:margin=2
10839 @section tinterlace
10841 Perform various types of temporal field interlacing.
10843 Frames are counted starting from 1, so the first input frame is
10846 The filter accepts the following options:
10851 Specify the mode of the interlacing. This option can also be specified
10852 as a value alone. See below for a list of values for this option.
10854 Available values are:
10858 Move odd frames into the upper field, even into the lower field,
10859 generating a double height frame at half frame rate.
10863 Frame 1 Frame 2 Frame 3 Frame 4
10865 11111 22222 33333 44444
10866 11111 22222 33333 44444
10867 11111 22222 33333 44444
10868 11111 22222 33333 44444
10882 Only output even frames, odd frames are dropped, generating a frame with
10883 unchanged height at half frame rate.
10888 Frame 1 Frame 2 Frame 3 Frame 4
10890 11111 22222 33333 44444
10891 11111 22222 33333 44444
10892 11111 22222 33333 44444
10893 11111 22222 33333 44444
10903 Only output odd frames, even frames are dropped, generating a frame with
10904 unchanged height at half frame rate.
10909 Frame 1 Frame 2 Frame 3 Frame 4
10911 11111 22222 33333 44444
10912 11111 22222 33333 44444
10913 11111 22222 33333 44444
10914 11111 22222 33333 44444
10924 Expand each frame to full height, but pad alternate lines with black,
10925 generating a frame with double height at the same input frame rate.
10930 Frame 1 Frame 2 Frame 3 Frame 4
10932 11111 22222 33333 44444
10933 11111 22222 33333 44444
10934 11111 22222 33333 44444
10935 11111 22222 33333 44444
10938 11111 ..... 33333 .....
10939 ..... 22222 ..... 44444
10940 11111 ..... 33333 .....
10941 ..... 22222 ..... 44444
10942 11111 ..... 33333 .....
10943 ..... 22222 ..... 44444
10944 11111 ..... 33333 .....
10945 ..... 22222 ..... 44444
10949 @item interleave_top, 4
10950 Interleave the upper field from odd frames with the lower field from
10951 even frames, generating a frame with unchanged height at half frame rate.
10956 Frame 1 Frame 2 Frame 3 Frame 4
10958 11111<- 22222 33333<- 44444
10959 11111 22222<- 33333 44444<-
10960 11111<- 22222 33333<- 44444
10961 11111 22222<- 33333 44444<-
10971 @item interleave_bottom, 5
10972 Interleave the lower field from odd frames with the upper field from
10973 even frames, generating a frame with unchanged height at half frame rate.
10978 Frame 1 Frame 2 Frame 3 Frame 4
10980 11111 22222<- 33333 44444<-
10981 11111<- 22222 33333<- 44444
10982 11111 22222<- 33333 44444<-
10983 11111<- 22222 33333<- 44444
10993 @item interlacex2, 6
10994 Double frame rate with unchanged height. Frames are inserted each
10995 containing the second temporal field from the previous input frame and
10996 the first temporal field from the next input frame. This mode relies on
10997 the top_field_first flag. Useful for interlaced video displays with no
10998 field synchronisation.
11003 Frame 1 Frame 2 Frame 3 Frame 4
11005 11111 22222 33333 44444
11006 11111 22222 33333 44444
11007 11111 22222 33333 44444
11008 11111 22222 33333 44444
11011 11111 22222 22222 33333 33333 44444 44444
11012 11111 11111 22222 22222 33333 33333 44444
11013 11111 22222 22222 33333 33333 44444 44444
11014 11111 11111 22222 22222 33333 33333 44444
11018 Move odd frames into the upper field, even into the lower field,
11019 generating a double height frame at same frame rate.
11023 Frame 1 Frame 2 Frame 3 Frame 4
11025 11111 22222 33333 44444
11026 11111 22222 33333 44444
11027 11111 22222 33333 44444
11028 11111 22222 33333 44444
11031 11111 33333 33333 55555
11032 22222 22222 44444 44444
11033 11111 33333 33333 55555
11034 22222 22222 44444 44444
11035 11111 33333 33333 55555
11036 22222 22222 44444 44444
11037 11111 33333 33333 55555
11038 22222 22222 44444 44444
11043 Numeric values are deprecated but are accepted for backward
11044 compatibility reasons.
11046 Default mode is @code{merge}.
11049 Specify flags influencing the filter process.
11051 Available value for @var{flags} is:
11054 @item low_pass_filter, vlfp
11055 Enable vertical low-pass filtering in the filter.
11056 Vertical low-pass filtering is required when creating an interlaced
11057 destination from a progressive source which contains high-frequency
11058 vertical detail. Filtering will reduce interlace 'twitter' and Moire
11061 Vertical low-pass filtering can only be enabled for @option{mode}
11062 @var{interleave_top} and @var{interleave_bottom}.
11069 Transpose rows with columns in the input video and optionally flip it.
11071 It accepts the following parameters:
11076 Specify the transposition direction.
11078 Can assume the following values:
11080 @item 0, 4, cclock_flip
11081 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
11089 Rotate by 90 degrees clockwise, that is:
11097 Rotate by 90 degrees counterclockwise, that is:
11104 @item 3, 7, clock_flip
11105 Rotate by 90 degrees clockwise and vertically flip, that is:
11113 For values between 4-7, the transposition is only done if the input
11114 video geometry is portrait and not landscape. These values are
11115 deprecated, the @code{passthrough} option should be used instead.
11117 Numerical values are deprecated, and should be dropped in favor of
11118 symbolic constants.
11121 Do not apply the transposition if the input geometry matches the one
11122 specified by the specified value. It accepts the following values:
11125 Always apply transposition.
11127 Preserve portrait geometry (when @var{height} >= @var{width}).
11129 Preserve landscape geometry (when @var{width} >= @var{height}).
11132 Default value is @code{none}.
11135 For example to rotate by 90 degrees clockwise and preserve portrait
11138 transpose=dir=1:passthrough=portrait
11141 The command above can also be specified as:
11143 transpose=1:portrait
11147 Trim the input so that the output contains one continuous subpart of the input.
11149 It accepts the following parameters:
11152 Specify the time of the start of the kept section, i.e. the frame with the
11153 timestamp @var{start} will be the first frame in the output.
11156 Specify the time of the first frame that will be dropped, i.e. the frame
11157 immediately preceding the one with the timestamp @var{end} will be the last
11158 frame in the output.
11161 This is the same as @var{start}, except this option sets the start timestamp
11162 in timebase units instead of seconds.
11165 This is the same as @var{end}, except this option sets the end timestamp
11166 in timebase units instead of seconds.
11169 The maximum duration of the output in seconds.
11172 The number of the first frame that should be passed to the output.
11175 The number of the first frame that should be dropped.
11178 @option{start}, @option{end}, and @option{duration} are expressed as time
11179 duration specifications; see
11180 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
11181 for the accepted syntax.
11183 Note that the first two sets of the start/end options and the @option{duration}
11184 option look at the frame timestamp, while the _frame variants simply count the
11185 frames that pass through the filter. Also note that this filter does not modify
11186 the timestamps. If you wish for the output timestamps to start at zero, insert a
11187 setpts filter after the trim filter.
11189 If multiple start or end options are set, this filter tries to be greedy and
11190 keep all the frames that match at least one of the specified constraints. To keep
11191 only the part that matches all the constraints at once, chain multiple trim
11194 The defaults are such that all the input is kept. So it is possible to set e.g.
11195 just the end values to keep everything before the specified time.
11200 Drop everything except the second minute of input:
11202 ffmpeg -i INPUT -vf trim=60:120
11206 Keep only the first second:
11208 ffmpeg -i INPUT -vf trim=duration=1
11217 Sharpen or blur the input video.
11219 It accepts the following parameters:
11222 @item luma_msize_x, lx
11223 Set the luma matrix horizontal size. It must be an odd integer between
11224 3 and 63. The default value is 5.
11226 @item luma_msize_y, ly
11227 Set the luma matrix vertical size. It must be an odd integer between 3
11228 and 63. The default value is 5.
11230 @item luma_amount, la
11231 Set the luma effect strength. It must be a floating point number, reasonable
11232 values lay between -1.5 and 1.5.
11234 Negative values will blur the input video, while positive values will
11235 sharpen it, a value of zero will disable the effect.
11237 Default value is 1.0.
11239 @item chroma_msize_x, cx
11240 Set the chroma matrix horizontal size. It must be an odd integer
11241 between 3 and 63. The default value is 5.
11243 @item chroma_msize_y, cy
11244 Set the chroma matrix vertical size. It must be an odd integer
11245 between 3 and 63. The default value is 5.
11247 @item chroma_amount, ca
11248 Set the chroma effect strength. It must be a floating point number, reasonable
11249 values lay between -1.5 and 1.5.
11251 Negative values will blur the input video, while positive values will
11252 sharpen it, a value of zero will disable the effect.
11254 Default value is 0.0.
11257 If set to 1, specify using OpenCL capabilities, only available if
11258 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
11262 All parameters are optional and default to the equivalent of the
11263 string '5:5:1.0:5:5:0.0'.
11265 @subsection Examples
11269 Apply strong luma sharpen effect:
11271 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
11275 Apply a strong blur of both luma and chroma parameters:
11277 unsharp=7:7:-2:7:7:-2
11283 Apply ultra slow/simple postprocessing filter that compresses and decompresses
11284 the image at several (or - in the case of @option{quality} level @code{8} - all)
11285 shifts and average the results.
11287 The way this differs from the behavior of spp is that uspp actually encodes &
11288 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
11289 DCT similar to MJPEG.
11291 The filter accepts the following options:
11295 Set quality. This option defines the number of levels for averaging. It accepts
11296 an integer in the range 0-8. If set to @code{0}, the filter will have no
11297 effect. A value of @code{8} means the higher quality. For each increment of
11298 that value the speed drops by a factor of approximately 2. Default value is
11302 Force a constant quantization parameter. If not set, the filter will use the QP
11303 from the video stream (if available).
11306 @section vectorscope
11308 Display 2 color component values in the two dimensional graph (which is called
11311 This filter accepts the following options:
11315 Set vectorscope mode.
11317 It accepts the following values:
11320 Gray values are displayed on graph, higher brightness means more pixels have
11321 same component color value on location in graph. This is the default mode.
11324 Gray values are displayed on graph. Surrounding pixels values which are not
11325 present in video frame are drawn in gradient of 2 color components which are
11326 set by option @code{x} and @code{y}.
11329 Actual color components values present in video frame are displayed on graph.
11332 Similar as color2 but higher frequency of same values @code{x} and @code{y}
11333 on graph increases value of another color component, which is luminance by
11334 default values of @code{x} and @code{y}.
11337 Actual colors present in video frame are displayed on graph. If two different
11338 colors map to same position on graph then color with higher value of component
11339 not present in graph is picked.
11343 Set which color component will be represented on X-axis. Default is @code{1}.
11346 Set which color component will be represented on Y-axis. Default is @code{2}.
11349 Set intensity, used by modes: gray, color and color3 for increasing brightness
11350 of color component which represents frequency of (X, Y) location in graph.
11355 No envelope, this is default.
11358 Instant envelope, even darkest single pixel will be clearly highlighted.
11361 Hold maximum and minimum values presented in graph over time. This way you
11362 can still spot out of range values without constantly looking at vectorscope.
11365 Peak and instant envelope combined together.
11369 @anchor{vidstabdetect}
11370 @section vidstabdetect
11372 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
11373 @ref{vidstabtransform} for pass 2.
11375 This filter generates a file with relative translation and rotation
11376 transform information about subsequent frames, which is then used by
11377 the @ref{vidstabtransform} filter.
11379 To enable compilation of this filter you need to configure FFmpeg with
11380 @code{--enable-libvidstab}.
11382 This filter accepts the following options:
11386 Set the path to the file used to write the transforms information.
11387 Default value is @file{transforms.trf}.
11390 Set how shaky the video is and how quick the camera is. It accepts an
11391 integer in the range 1-10, a value of 1 means little shakiness, a
11392 value of 10 means strong shakiness. Default value is 5.
11395 Set the accuracy of the detection process. It must be a value in the
11396 range 1-15. A value of 1 means low accuracy, a value of 15 means high
11397 accuracy. Default value is 15.
11400 Set stepsize of the search process. The region around minimum is
11401 scanned with 1 pixel resolution. Default value is 6.
11404 Set minimum contrast. Below this value a local measurement field is
11405 discarded. Must be a floating point value in the range 0-1. Default
11409 Set reference frame number for tripod mode.
11411 If enabled, the motion of the frames is compared to a reference frame
11412 in the filtered stream, identified by the specified number. The idea
11413 is to compensate all movements in a more-or-less static scene and keep
11414 the camera view absolutely still.
11416 If set to 0, it is disabled. The frames are counted starting from 1.
11419 Show fields and transforms in the resulting frames. It accepts an
11420 integer in the range 0-2. Default value is 0, which disables any
11424 @subsection Examples
11428 Use default values:
11434 Analyze strongly shaky movie and put the results in file
11435 @file{mytransforms.trf}:
11437 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
11441 Visualize the result of internal transformations in the resulting
11444 vidstabdetect=show=1
11448 Analyze a video with medium shakiness using @command{ffmpeg}:
11450 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
11454 @anchor{vidstabtransform}
11455 @section vidstabtransform
11457 Video stabilization/deshaking: pass 2 of 2,
11458 see @ref{vidstabdetect} for pass 1.
11460 Read a file with transform information for each frame and
11461 apply/compensate them. Together with the @ref{vidstabdetect}
11462 filter this can be used to deshake videos. See also
11463 @url{http://public.hronopik.de/vid.stab}. It is important to also use
11464 the @ref{unsharp} filter, see below.
11466 To enable compilation of this filter you need to configure FFmpeg with
11467 @code{--enable-libvidstab}.
11469 @subsection Options
11473 Set path to the file used to read the transforms. Default value is
11474 @file{transforms.trf}.
11477 Set the number of frames (value*2 + 1) used for lowpass filtering the
11478 camera movements. Default value is 10.
11480 For example a number of 10 means that 21 frames are used (10 in the
11481 past and 10 in the future) to smoothen the motion in the video. A
11482 larger value leads to a smoother video, but limits the acceleration of
11483 the camera (pan/tilt movements). 0 is a special case where a static
11484 camera is simulated.
11487 Set the camera path optimization algorithm.
11489 Accepted values are:
11492 gaussian kernel low-pass filter on camera motion (default)
11494 averaging on transformations
11498 Set maximal number of pixels to translate frames. Default value is -1,
11502 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
11503 value is -1, meaning no limit.
11506 Specify how to deal with borders that may be visible due to movement
11509 Available values are:
11512 keep image information from previous frame (default)
11514 fill the border black
11518 Invert transforms if set to 1. Default value is 0.
11521 Consider transforms as relative to previous frame if set to 1,
11522 absolute if set to 0. Default value is 0.
11525 Set percentage to zoom. A positive value will result in a zoom-in
11526 effect, a negative value in a zoom-out effect. Default value is 0 (no
11530 Set optimal zooming to avoid borders.
11532 Accepted values are:
11537 optimal static zoom value is determined (only very strong movements
11538 will lead to visible borders) (default)
11540 optimal adaptive zoom value is determined (no borders will be
11541 visible), see @option{zoomspeed}
11544 Note that the value given at zoom is added to the one calculated here.
11547 Set percent to zoom maximally each frame (enabled when
11548 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
11552 Specify type of interpolation.
11554 Available values are:
11559 linear only horizontal
11561 linear in both directions (default)
11563 cubic in both directions (slow)
11567 Enable virtual tripod mode if set to 1, which is equivalent to
11568 @code{relative=0:smoothing=0}. Default value is 0.
11570 Use also @code{tripod} option of @ref{vidstabdetect}.
11573 Increase log verbosity if set to 1. Also the detected global motions
11574 are written to the temporary file @file{global_motions.trf}. Default
11578 @subsection Examples
11582 Use @command{ffmpeg} for a typical stabilization with default values:
11584 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
11587 Note the use of the @ref{unsharp} filter which is always recommended.
11590 Zoom in a bit more and load transform data from a given file:
11592 vidstabtransform=zoom=5:input="mytransforms.trf"
11596 Smoothen the video even more:
11598 vidstabtransform=smoothing=30
11604 Flip the input video vertically.
11606 For example, to vertically flip a video with @command{ffmpeg}:
11608 ffmpeg -i in.avi -vf "vflip" out.avi
11614 Make or reverse a natural vignetting effect.
11616 The filter accepts the following options:
11620 Set lens angle expression as a number of radians.
11622 The value is clipped in the @code{[0,PI/2]} range.
11624 Default value: @code{"PI/5"}
11628 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
11632 Set forward/backward mode.
11634 Available modes are:
11637 The larger the distance from the central point, the darker the image becomes.
11640 The larger the distance from the central point, the brighter the image becomes.
11641 This can be used to reverse a vignette effect, though there is no automatic
11642 detection to extract the lens @option{angle} and other settings (yet). It can
11643 also be used to create a burning effect.
11646 Default value is @samp{forward}.
11649 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
11651 It accepts the following values:
11654 Evaluate expressions only once during the filter initialization.
11657 Evaluate expressions for each incoming frame. This is way slower than the
11658 @samp{init} mode since it requires all the scalers to be re-computed, but it
11659 allows advanced dynamic expressions.
11662 Default value is @samp{init}.
11665 Set dithering to reduce the circular banding effects. Default is @code{1}
11669 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
11670 Setting this value to the SAR of the input will make a rectangular vignetting
11671 following the dimensions of the video.
11673 Default is @code{1/1}.
11676 @subsection Expressions
11678 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
11679 following parameters.
11684 input width and height
11687 the number of input frame, starting from 0
11690 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
11691 @var{TB} units, NAN if undefined
11694 frame rate of the input video, NAN if the input frame rate is unknown
11697 the PTS (Presentation TimeStamp) of the filtered video frame,
11698 expressed in seconds, NAN if undefined
11701 time base of the input video
11705 @subsection Examples
11709 Apply simple strong vignetting effect:
11715 Make a flickering vignetting:
11717 vignette='PI/4+random(1)*PI/50':eval=frame
11723 Stack input videos vertically.
11725 All streams must be of same pixel format and of same width.
11727 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
11728 to create same output.
11730 The filter accept the following option:
11734 Set number of input streams. Default is 2.
11737 If set to 1, force the output to terminate when the shortest input
11738 terminates. Default value is 0.
11743 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
11744 Deinterlacing Filter").
11746 Based on the process described by Martin Weston for BBC R&D, and
11747 implemented based on the de-interlace algorithm written by Jim
11748 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
11749 uses filter coefficients calculated by BBC R&D.
11751 There are two sets of filter coefficients, so called "simple":
11752 and "complex". Which set of filter coefficients is used can
11753 be set by passing an optional parameter:
11757 Set the interlacing filter coefficients. Accepts one of the following values:
11761 Simple filter coefficient set.
11763 More-complex filter coefficient set.
11765 Default value is @samp{complex}.
11768 Specify which frames to deinterlace. Accept one of the following values:
11772 Deinterlace all frames,
11774 Only deinterlace frames marked as interlaced.
11777 Default value is @samp{all}.
11781 Video waveform monitor.
11783 The waveform monitor plots color component intensity. By default luminance
11784 only. Each column of the waveform corresponds to a column of pixels in the
11787 It accepts the following options:
11791 Can be either @code{row}, or @code{column}. Default is @code{column}.
11792 In row mode, the graph on the left side represents color component value 0 and
11793 the right side represents value = 255. In column mode, the top side represents
11794 color component value = 0 and bottom side represents value = 255.
11797 Set intensity. Smaller values are useful to find out how many values of the same
11798 luminance are distributed across input rows/columns.
11799 Default value is @code{0.04}. Allowed range is [0, 1].
11802 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
11803 In mirrored mode, higher values will be represented on the left
11804 side for @code{row} mode and at the top for @code{column} mode. Default is
11805 @code{1} (mirrored).
11809 It accepts the following values:
11812 Presents information identical to that in the @code{parade}, except
11813 that the graphs representing color components are superimposed directly
11816 This display mode makes it easier to spot relative differences or similarities
11817 in overlapping areas of the color components that are supposed to be identical,
11818 such as neutral whites, grays, or blacks.
11821 Display separate graph for the color components side by side in
11822 @code{row} mode or one below the other in @code{column} mode.
11824 Using this display mode makes it easy to spot color casts in the highlights
11825 and shadows of an image, by comparing the contours of the top and the bottom
11826 graphs of each waveform. Since whites, grays, and blacks are characterized
11827 by exactly equal amounts of red, green, and blue, neutral areas of the picture
11828 should display three waveforms of roughly equal width/height. If not, the
11829 correction is easy to perform by making level adjustments the three waveforms.
11831 Default is @code{parade}.
11833 @item components, c
11834 Set which color components to display. Default is 1, which means only luminance
11835 or red color component if input is in RGB colorspace. If is set for example to
11836 7 it will display all 3 (if) available color components.
11841 No envelope, this is default.
11844 Instant envelope, minimum and maximum values presented in graph will be easily
11845 visible even with small @code{step} value.
11848 Hold minimum and maximum values presented in graph across time. This way you
11849 can still spot out of range values without constantly looking at waveforms.
11852 Peak and instant envelope combined together.
11858 No filtering, this is default.
11861 Luma and chroma combined together.
11864 Similar as above, but shows difference between blue and red chroma.
11867 Displays only chroma.
11870 Similar as above, but shows difference between blue and red chroma.
11873 Displays actual color value on waveform.
11878 Apply the xBR high-quality magnification filter which is designed for pixel
11879 art. It follows a set of edge-detection rules, see
11880 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
11882 It accepts the following option:
11886 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
11887 @code{3xBR} and @code{4} for @code{4xBR}.
11888 Default is @code{3}.
11894 Deinterlace the input video ("yadif" means "yet another deinterlacing
11897 It accepts the following parameters:
11903 The interlacing mode to adopt. It accepts one of the following values:
11906 @item 0, send_frame
11907 Output one frame for each frame.
11908 @item 1, send_field
11909 Output one frame for each field.
11910 @item 2, send_frame_nospatial
11911 Like @code{send_frame}, but it skips the spatial interlacing check.
11912 @item 3, send_field_nospatial
11913 Like @code{send_field}, but it skips the spatial interlacing check.
11916 The default value is @code{send_frame}.
11919 The picture field parity assumed for the input interlaced video. It accepts one
11920 of the following values:
11924 Assume the top field is first.
11926 Assume the bottom field is first.
11928 Enable automatic detection of field parity.
11931 The default value is @code{auto}.
11932 If the interlacing is unknown or the decoder does not export this information,
11933 top field first will be assumed.
11936 Specify which frames to deinterlace. Accept one of the following
11941 Deinterlace all frames.
11942 @item 1, interlaced
11943 Only deinterlace frames marked as interlaced.
11946 The default value is @code{all}.
11951 Apply Zoom & Pan effect.
11953 This filter accepts the following options:
11957 Set the zoom expression. Default is 1.
11961 Set the x and y expression. Default is 0.
11964 Set the duration expression in number of frames.
11965 This sets for how many number of frames effect will last for
11966 single input image.
11969 Set the output image size, default is 'hd720'.
11972 Each expression can contain the following constants:
11991 Output frame count.
11995 Last calculated 'x' and 'y' position from 'x' and 'y' expression
11996 for current input frame.
12000 'x' and 'y' of last output frame of previous input frame or 0 when there was
12001 not yet such frame (first input frame).
12004 Last calculated zoom from 'z' expression for current input frame.
12007 Last calculated zoom of last output frame of previous input frame.
12010 Number of output frames for current input frame. Calculated from 'd' expression
12011 for each input frame.
12014 number of output frames created for previous input frame
12017 Rational number: input width / input height
12020 sample aspect ratio
12023 display aspect ratio
12027 @subsection Examples
12031 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
12033 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
12037 Zoom-in up to 1.5 and pan always at center of picture:
12039 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
12044 Scale (resize) the input video, using the z.lib library:
12045 https://github.com/sekrit-twc/zimg.
12047 The zscale filter forces the output display aspect ratio to be the same
12048 as the input, by changing the output sample aspect ratio.
12050 If the input image format is different from the format requested by
12051 the next filter, the zscale filter will convert the input to the
12054 @subsection Options
12055 The filter accepts the following options.
12060 Set the output video dimension expression. Default value is the input
12063 If the @var{width} or @var{w} is 0, the input width is used for the output.
12064 If the @var{height} or @var{h} is 0, the input height is used for the output.
12066 If one of the values is -1, the zscale filter will use a value that
12067 maintains the aspect ratio of the input image, calculated from the
12068 other specified dimension. If both of them are -1, the input size is
12071 If one of the values is -n with n > 1, the zscale filter will also use a value
12072 that maintains the aspect ratio of the input image, calculated from the other
12073 specified dimension. After that it will, however, make sure that the calculated
12074 dimension is divisible by n and adjust the value if necessary.
12076 See below for the list of accepted constants for use in the dimension
12080 Set the video size. For the syntax of this option, check the
12081 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12084 Set the dither type.
12086 Possible values are:
12091 @item error_diffusion
12097 Set the resize filter type.
12099 Possible values are:
12109 Default is bilinear.
12112 Set the color range.
12114 Possible values are:
12121 Default is same as input.
12124 Set the color primaries.
12126 Possible values are:
12136 Default is same as input.
12139 Set the transfer characteristics.
12141 Possible values are:
12152 Default is same as input.
12155 Set the colorspace matrix.
12157 Possible value are:
12168 Default is same as input.
12171 The values of the @option{w} and @option{h} options are expressions
12172 containing the following constants:
12177 The input width and height
12181 These are the same as @var{in_w} and @var{in_h}.
12185 The output (scaled) width and height
12189 These are the same as @var{out_w} and @var{out_h}
12192 The same as @var{iw} / @var{ih}
12195 input sample aspect ratio
12198 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
12202 horizontal and vertical input chroma subsample values. For example for the
12203 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12207 horizontal and vertical output chroma subsample values. For example for the
12208 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12214 @c man end VIDEO FILTERS
12216 @chapter Video Sources
12217 @c man begin VIDEO SOURCES
12219 Below is a description of the currently available video sources.
12223 Buffer video frames, and make them available to the filter chain.
12225 This source is mainly intended for a programmatic use, in particular
12226 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
12228 It accepts the following parameters:
12233 Specify the size (width and height) of the buffered video frames. For the
12234 syntax of this option, check the
12235 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12238 The input video width.
12241 The input video height.
12244 A string representing the pixel format of the buffered video frames.
12245 It may be a number corresponding to a pixel format, or a pixel format
12249 Specify the timebase assumed by the timestamps of the buffered frames.
12252 Specify the frame rate expected for the video stream.
12254 @item pixel_aspect, sar
12255 The sample (pixel) aspect ratio of the input video.
12258 Specify the optional parameters to be used for the scale filter which
12259 is automatically inserted when an input change is detected in the
12260 input size or format.
12265 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
12268 will instruct the source to accept video frames with size 320x240 and
12269 with format "yuv410p", assuming 1/24 as the timestamps timebase and
12270 square pixels (1:1 sample aspect ratio).
12271 Since the pixel format with name "yuv410p" corresponds to the number 6
12272 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
12273 this example corresponds to:
12275 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
12278 Alternatively, the options can be specified as a flat string, but this
12279 syntax is deprecated:
12281 @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}]
12285 Create a pattern generated by an elementary cellular automaton.
12287 The initial state of the cellular automaton can be defined through the
12288 @option{filename}, and @option{pattern} options. If such options are
12289 not specified an initial state is created randomly.
12291 At each new frame a new row in the video is filled with the result of
12292 the cellular automaton next generation. The behavior when the whole
12293 frame is filled is defined by the @option{scroll} option.
12295 This source accepts the following options:
12299 Read the initial cellular automaton state, i.e. the starting row, from
12300 the specified file.
12301 In the file, each non-whitespace character is considered an alive
12302 cell, a newline will terminate the row, and further characters in the
12303 file will be ignored.
12306 Read the initial cellular automaton state, i.e. the starting row, from
12307 the specified string.
12309 Each non-whitespace character in the string is considered an alive
12310 cell, a newline will terminate the row, and further characters in the
12311 string will be ignored.
12314 Set the video rate, that is the number of frames generated per second.
12317 @item random_fill_ratio, ratio
12318 Set the random fill ratio for the initial cellular automaton row. It
12319 is a floating point number value ranging from 0 to 1, defaults to
12322 This option is ignored when a file or a pattern is specified.
12324 @item random_seed, seed
12325 Set the seed for filling randomly the initial row, must be an integer
12326 included between 0 and UINT32_MAX. If not specified, or if explicitly
12327 set to -1, the filter will try to use a good random seed on a best
12331 Set the cellular automaton rule, it is a number ranging from 0 to 255.
12332 Default value is 110.
12335 Set the size of the output video. For the syntax of this option, check the
12336 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12338 If @option{filename} or @option{pattern} is specified, the size is set
12339 by default to the width of the specified initial state row, and the
12340 height is set to @var{width} * PHI.
12342 If @option{size} is set, it must contain the width of the specified
12343 pattern string, and the specified pattern will be centered in the
12346 If a filename or a pattern string is not specified, the size value
12347 defaults to "320x518" (used for a randomly generated initial state).
12350 If set to 1, scroll the output upward when all the rows in the output
12351 have been already filled. If set to 0, the new generated row will be
12352 written over the top row just after the bottom row is filled.
12355 @item start_full, full
12356 If set to 1, completely fill the output with generated rows before
12357 outputting the first frame.
12358 This is the default behavior, for disabling set the value to 0.
12361 If set to 1, stitch the left and right row edges together.
12362 This is the default behavior, for disabling set the value to 0.
12365 @subsection Examples
12369 Read the initial state from @file{pattern}, and specify an output of
12372 cellauto=f=pattern:s=200x400
12376 Generate a random initial row with a width of 200 cells, with a fill
12379 cellauto=ratio=2/3:s=200x200
12383 Create a pattern generated by rule 18 starting by a single alive cell
12384 centered on an initial row with width 100:
12386 cellauto=p=@@:s=100x400:full=0:rule=18
12390 Specify a more elaborated initial pattern:
12392 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
12397 @section mandelbrot
12399 Generate a Mandelbrot set fractal, and progressively zoom towards the
12400 point specified with @var{start_x} and @var{start_y}.
12402 This source accepts the following options:
12407 Set the terminal pts value. Default value is 400.
12410 Set the terminal scale value.
12411 Must be a floating point value. Default value is 0.3.
12414 Set the inner coloring mode, that is the algorithm used to draw the
12415 Mandelbrot fractal internal region.
12417 It shall assume one of the following values:
12422 Show time until convergence.
12424 Set color based on point closest to the origin of the iterations.
12429 Default value is @var{mincol}.
12432 Set the bailout value. Default value is 10.0.
12435 Set the maximum of iterations performed by the rendering
12436 algorithm. Default value is 7189.
12439 Set outer coloring mode.
12440 It shall assume one of following values:
12442 @item iteration_count
12443 Set iteration cound mode.
12444 @item normalized_iteration_count
12445 set normalized iteration count mode.
12447 Default value is @var{normalized_iteration_count}.
12450 Set frame rate, expressed as number of frames per second. Default
12454 Set frame size. For the syntax of this option, check the "Video
12455 size" section in the ffmpeg-utils manual. Default value is "640x480".
12458 Set the initial scale value. Default value is 3.0.
12461 Set the initial x position. Must be a floating point value between
12462 -100 and 100. Default value is -0.743643887037158704752191506114774.
12465 Set the initial y position. Must be a floating point value between
12466 -100 and 100. Default value is -0.131825904205311970493132056385139.
12471 Generate various test patterns, as generated by the MPlayer test filter.
12473 The size of the generated video is fixed, and is 256x256.
12474 This source is useful in particular for testing encoding features.
12476 This source accepts the following options:
12481 Specify the frame rate of the sourced video, as the number of frames
12482 generated per second. It has to be a string in the format
12483 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
12484 number or a valid video frame rate abbreviation. The default value is
12488 Set the duration of the sourced video. See
12489 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12490 for the accepted syntax.
12492 If not specified, or the expressed duration is negative, the video is
12493 supposed to be generated forever.
12497 Set the number or the name of the test to perform. Supported tests are:
12513 Default value is "all", which will cycle through the list of all tests.
12518 mptestsrc=t=dc_luma
12521 will generate a "dc_luma" test pattern.
12523 @section frei0r_src
12525 Provide a frei0r source.
12527 To enable compilation of this filter you need to install the frei0r
12528 header and configure FFmpeg with @code{--enable-frei0r}.
12530 This source accepts the following parameters:
12535 The size of the video to generate. For the syntax of this option, check the
12536 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12539 The framerate of the generated video. It may be a string of the form
12540 @var{num}/@var{den} or a frame rate abbreviation.
12543 The name to the frei0r source to load. For more information regarding frei0r and
12544 how to set the parameters, read the @ref{frei0r} section in the video filters
12547 @item filter_params
12548 A '|'-separated list of parameters to pass to the frei0r source.
12552 For example, to generate a frei0r partik0l source with size 200x200
12553 and frame rate 10 which is overlaid on the overlay filter main input:
12555 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
12560 Generate a life pattern.
12562 This source is based on a generalization of John Conway's life game.
12564 The sourced input represents a life grid, each pixel represents a cell
12565 which can be in one of two possible states, alive or dead. Every cell
12566 interacts with its eight neighbours, which are the cells that are
12567 horizontally, vertically, or diagonally adjacent.
12569 At each interaction the grid evolves according to the adopted rule,
12570 which specifies the number of neighbor alive cells which will make a
12571 cell stay alive or born. The @option{rule} option allows one to specify
12574 This source accepts the following options:
12578 Set the file from which to read the initial grid state. In the file,
12579 each non-whitespace character is considered an alive cell, and newline
12580 is used to delimit the end of each row.
12582 If this option is not specified, the initial grid is generated
12586 Set the video rate, that is the number of frames generated per second.
12589 @item random_fill_ratio, ratio
12590 Set the random fill ratio for the initial random grid. It is a
12591 floating point number value ranging from 0 to 1, defaults to 1/PHI.
12592 It is ignored when a file is specified.
12594 @item random_seed, seed
12595 Set the seed for filling the initial random grid, must be an integer
12596 included between 0 and UINT32_MAX. If not specified, or if explicitly
12597 set to -1, the filter will try to use a good random seed on a best
12603 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
12604 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
12605 @var{NS} specifies the number of alive neighbor cells which make a
12606 live cell stay alive, and @var{NB} the number of alive neighbor cells
12607 which make a dead cell to become alive (i.e. to "born").
12608 "s" and "b" can be used in place of "S" and "B", respectively.
12610 Alternatively a rule can be specified by an 18-bits integer. The 9
12611 high order bits are used to encode the next cell state if it is alive
12612 for each number of neighbor alive cells, the low order bits specify
12613 the rule for "borning" new cells. Higher order bits encode for an
12614 higher number of neighbor cells.
12615 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
12616 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
12618 Default value is "S23/B3", which is the original Conway's game of life
12619 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
12620 cells, and will born a new cell if there are three alive cells around
12624 Set the size of the output video. For the syntax of this option, check the
12625 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12627 If @option{filename} is specified, the size is set by default to the
12628 same size of the input file. If @option{size} is set, it must contain
12629 the size specified in the input file, and the initial grid defined in
12630 that file is centered in the larger resulting area.
12632 If a filename is not specified, the size value defaults to "320x240"
12633 (used for a randomly generated initial grid).
12636 If set to 1, stitch the left and right grid edges together, and the
12637 top and bottom edges also. Defaults to 1.
12640 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
12641 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
12642 value from 0 to 255.
12645 Set the color of living (or new born) cells.
12648 Set the color of dead cells. If @option{mold} is set, this is the first color
12649 used to represent a dead cell.
12652 Set mold color, for definitely dead and moldy cells.
12654 For the syntax of these 3 color options, check the "Color" section in the
12655 ffmpeg-utils manual.
12658 @subsection Examples
12662 Read a grid from @file{pattern}, and center it on a grid of size
12665 life=f=pattern:s=300x300
12669 Generate a random grid of size 200x200, with a fill ratio of 2/3:
12671 life=ratio=2/3:s=200x200
12675 Specify a custom rule for evolving a randomly generated grid:
12681 Full example with slow death effect (mold) using @command{ffplay}:
12683 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
12690 @anchor{haldclutsrc}
12692 @anchor{rgbtestsrc}
12694 @anchor{smptehdbars}
12696 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
12698 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
12700 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
12702 The @code{color} source provides an uniformly colored input.
12704 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
12705 @ref{haldclut} filter.
12707 The @code{nullsrc} source returns unprocessed video frames. It is
12708 mainly useful to be employed in analysis / debugging tools, or as the
12709 source for filters which ignore the input data.
12711 The @code{rgbtestsrc} source generates an RGB test pattern useful for
12712 detecting RGB vs BGR issues. You should see a red, green and blue
12713 stripe from top to bottom.
12715 The @code{smptebars} source generates a color bars pattern, based on
12716 the SMPTE Engineering Guideline EG 1-1990.
12718 The @code{smptehdbars} source generates a color bars pattern, based on
12719 the SMPTE RP 219-2002.
12721 The @code{testsrc} source generates a test video pattern, showing a
12722 color pattern, a scrolling gradient and a timestamp. This is mainly
12723 intended for testing purposes.
12725 The sources accept the following parameters:
12730 Specify the color of the source, only available in the @code{color}
12731 source. For the syntax of this option, check the "Color" section in the
12732 ffmpeg-utils manual.
12735 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
12736 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
12737 pixels to be used as identity matrix for 3D lookup tables. Each component is
12738 coded on a @code{1/(N*N)} scale.
12741 Specify the size of the sourced video. For the syntax of this option, check the
12742 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12743 The default value is @code{320x240}.
12745 This option is not available with the @code{haldclutsrc} filter.
12748 Specify the frame rate of the sourced video, as the number of frames
12749 generated per second. It has to be a string in the format
12750 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
12751 number or a valid video frame rate abbreviation. The default value is
12755 Set the sample aspect ratio of the sourced video.
12758 Set the duration of the sourced video. See
12759 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12760 for the accepted syntax.
12762 If not specified, or the expressed duration is negative, the video is
12763 supposed to be generated forever.
12766 Set the number of decimals to show in the timestamp, only available in the
12767 @code{testsrc} source.
12769 The displayed timestamp value will correspond to the original
12770 timestamp value multiplied by the power of 10 of the specified
12771 value. Default value is 0.
12774 For example the following:
12776 testsrc=duration=5.3:size=qcif:rate=10
12779 will generate a video with a duration of 5.3 seconds, with size
12780 176x144 and a frame rate of 10 frames per second.
12782 The following graph description will generate a red source
12783 with an opacity of 0.2, with size "qcif" and a frame rate of 10
12786 color=c=red@@0.2:s=qcif:r=10
12789 If the input content is to be ignored, @code{nullsrc} can be used. The
12790 following command generates noise in the luminance plane by employing
12791 the @code{geq} filter:
12793 nullsrc=s=256x256, geq=random(1)*255:128:128
12796 @subsection Commands
12798 The @code{color} source supports the following commands:
12802 Set the color of the created image. Accepts the same syntax of the
12803 corresponding @option{color} option.
12806 @c man end VIDEO SOURCES
12808 @chapter Video Sinks
12809 @c man begin VIDEO SINKS
12811 Below is a description of the currently available video sinks.
12813 @section buffersink
12815 Buffer video frames, and make them available to the end of the filter
12818 This sink is mainly intended for programmatic use, in particular
12819 through the interface defined in @file{libavfilter/buffersink.h}
12820 or the options system.
12822 It accepts a pointer to an AVBufferSinkContext structure, which
12823 defines the incoming buffers' formats, to be passed as the opaque
12824 parameter to @code{avfilter_init_filter} for initialization.
12828 Null video sink: do absolutely nothing with the input video. It is
12829 mainly useful as a template and for use in analysis / debugging
12832 @c man end VIDEO SINKS
12834 @chapter Multimedia Filters
12835 @c man begin MULTIMEDIA FILTERS
12837 Below is a description of the currently available multimedia filters.
12839 @section aphasemeter
12841 Convert input audio to a video output, displaying the audio phase.
12843 The filter accepts the following options:
12847 Set the output frame rate. Default value is @code{25}.
12850 Set the video size for the output. For the syntax of this option, check the
12851 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12852 Default value is @code{800x400}.
12857 Specify the red, green, blue contrast. Default values are @code{2},
12858 @code{7} and @code{1}.
12859 Allowed range is @code{[0, 255]}.
12862 Set color which will be used for drawing median phase. If color is
12863 @code{none} which is default, no median phase value will be drawn.
12866 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
12867 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
12868 The @code{-1} means left and right channels are completely out of phase and
12869 @code{1} means channels are in phase.
12871 @section avectorscope
12873 Convert input audio to a video output, representing the audio vector
12876 The filter is used to measure the difference between channels of stereo
12877 audio stream. A monoaural signal, consisting of identical left and right
12878 signal, results in straight vertical line. Any stereo separation is visible
12879 as a deviation from this line, creating a Lissajous figure.
12880 If the straight (or deviation from it) but horizontal line appears this
12881 indicates that the left and right channels are out of phase.
12883 The filter accepts the following options:
12887 Set the vectorscope mode.
12889 Available values are:
12892 Lissajous rotated by 45 degrees.
12895 Same as above but not rotated.
12898 Shape resembling half of circle.
12901 Default value is @samp{lissajous}.
12904 Set the video size for the output. For the syntax of this option, check the
12905 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12906 Default value is @code{400x400}.
12909 Set the output frame rate. Default value is @code{25}.
12915 Specify the red, green, blue and alpha contrast. Default values are @code{40},
12916 @code{160}, @code{80} and @code{255}.
12917 Allowed range is @code{[0, 255]}.
12923 Specify the red, green, blue and alpha fade. Default values are @code{15},
12924 @code{10}, @code{5} and @code{5}.
12925 Allowed range is @code{[0, 255]}.
12928 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
12931 @subsection Examples
12935 Complete example using @command{ffplay}:
12937 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
12938 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
12944 Concatenate audio and video streams, joining them together one after the
12947 The filter works on segments of synchronized video and audio streams. All
12948 segments must have the same number of streams of each type, and that will
12949 also be the number of streams at output.
12951 The filter accepts the following options:
12956 Set the number of segments. Default is 2.
12959 Set the number of output video streams, that is also the number of video
12960 streams in each segment. Default is 1.
12963 Set the number of output audio streams, that is also the number of audio
12964 streams in each segment. Default is 0.
12967 Activate unsafe mode: do not fail if segments have a different format.
12971 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
12972 @var{a} audio outputs.
12974 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
12975 segment, in the same order as the outputs, then the inputs for the second
12978 Related streams do not always have exactly the same duration, for various
12979 reasons including codec frame size or sloppy authoring. For that reason,
12980 related synchronized streams (e.g. a video and its audio track) should be
12981 concatenated at once. The concat filter will use the duration of the longest
12982 stream in each segment (except the last one), and if necessary pad shorter
12983 audio streams with silence.
12985 For this filter to work correctly, all segments must start at timestamp 0.
12987 All corresponding streams must have the same parameters in all segments; the
12988 filtering system will automatically select a common pixel format for video
12989 streams, and a common sample format, sample rate and channel layout for
12990 audio streams, but other settings, such as resolution, must be converted
12991 explicitly by the user.
12993 Different frame rates are acceptable but will result in variable frame rate
12994 at output; be sure to configure the output file to handle it.
12996 @subsection Examples
13000 Concatenate an opening, an episode and an ending, all in bilingual version
13001 (video in stream 0, audio in streams 1 and 2):
13003 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
13004 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
13005 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
13006 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
13010 Concatenate two parts, handling audio and video separately, using the
13011 (a)movie sources, and adjusting the resolution:
13013 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
13014 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
13015 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
13017 Note that a desync will happen at the stitch if the audio and video streams
13018 do not have exactly the same duration in the first file.
13025 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
13026 it unchanged. By default, it logs a message at a frequency of 10Hz with the
13027 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
13028 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
13030 The filter also has a video output (see the @var{video} option) with a real
13031 time graph to observe the loudness evolution. The graphic contains the logged
13032 message mentioned above, so it is not printed anymore when this option is set,
13033 unless the verbose logging is set. The main graphing area contains the
13034 short-term loudness (3 seconds of analysis), and the gauge on the right is for
13035 the momentary loudness (400 milliseconds).
13037 More information about the Loudness Recommendation EBU R128 on
13038 @url{http://tech.ebu.ch/loudness}.
13040 The filter accepts the following options:
13045 Activate the video output. The audio stream is passed unchanged whether this
13046 option is set or no. The video stream will be the first output stream if
13047 activated. Default is @code{0}.
13050 Set the video size. This option is for video only. For the syntax of this
13052 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13053 Default and minimum resolution is @code{640x480}.
13056 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
13057 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
13058 other integer value between this range is allowed.
13061 Set metadata injection. If set to @code{1}, the audio input will be segmented
13062 into 100ms output frames, each of them containing various loudness information
13063 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
13065 Default is @code{0}.
13068 Force the frame logging level.
13070 Available values are:
13073 information logging level
13075 verbose logging level
13078 By default, the logging level is set to @var{info}. If the @option{video} or
13079 the @option{metadata} options are set, it switches to @var{verbose}.
13084 Available modes can be cumulated (the option is a @code{flag} type). Possible
13088 Disable any peak mode (default).
13090 Enable sample-peak mode.
13092 Simple peak mode looking for the higher sample value. It logs a message
13093 for sample-peak (identified by @code{SPK}).
13095 Enable true-peak mode.
13097 If enabled, the peak lookup is done on an over-sampled version of the input
13098 stream for better peak accuracy. It logs a message for true-peak.
13099 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
13100 This mode requires a build with @code{libswresample}.
13104 Treat mono input files as "dual mono". If a mono file is intended for playback
13105 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
13106 If set to @code{true}, this option will compensate for this effect.
13107 Multi-channel input files are not affected by this option.
13110 Set a specific pan law to be used for the measurement of dual mono files.
13111 This parameter is optional, and has a default value of -3.01dB.
13114 @subsection Examples
13118 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
13120 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
13124 Run an analysis with @command{ffmpeg}:
13126 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
13130 @section interleave, ainterleave
13132 Temporally interleave frames from several inputs.
13134 @code{interleave} works with video inputs, @code{ainterleave} with audio.
13136 These filters read frames from several inputs and send the oldest
13137 queued frame to the output.
13139 Input streams must have a well defined, monotonically increasing frame
13142 In order to submit one frame to output, these filters need to enqueue
13143 at least one frame for each input, so they cannot work in case one
13144 input is not yet terminated and will not receive incoming frames.
13146 For example consider the case when one input is a @code{select} filter
13147 which always drop input frames. The @code{interleave} filter will keep
13148 reading from that input, but it will never be able to send new frames
13149 to output until the input will send an end-of-stream signal.
13151 Also, depending on inputs synchronization, the filters will drop
13152 frames in case one input receives more frames than the other ones, and
13153 the queue is already filled.
13155 These filters accept the following options:
13159 Set the number of different inputs, it is 2 by default.
13162 @subsection Examples
13166 Interleave frames belonging to different streams using @command{ffmpeg}:
13168 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
13172 Add flickering blur effect:
13174 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
13178 @section perms, aperms
13180 Set read/write permissions for the output frames.
13182 These filters are mainly aimed at developers to test direct path in the
13183 following filter in the filtergraph.
13185 The filters accept the following options:
13189 Select the permissions mode.
13191 It accepts the following values:
13194 Do nothing. This is the default.
13196 Set all the output frames read-only.
13198 Set all the output frames directly writable.
13200 Make the frame read-only if writable, and writable if read-only.
13202 Set each output frame read-only or writable randomly.
13206 Set the seed for the @var{random} mode, must be an integer included between
13207 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
13208 @code{-1}, the filter will try to use a good random seed on a best effort
13212 Note: in case of auto-inserted filter between the permission filter and the
13213 following one, the permission might not be received as expected in that
13214 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
13215 perms/aperms filter can avoid this problem.
13217 @section realtime, arealtime
13219 Slow down filtering to match real time approximatively.
13221 These filters will pause the filtering for a variable amount of time to
13222 match the output rate with the input timestamps.
13223 They are similar to the @option{re} option to @code{ffmpeg}.
13225 They accept the following options:
13229 Time limit for the pauses. Any pause longer than that will be considered
13230 a timestamp discontinuity and reset the timer. Default is 2 seconds.
13233 @section select, aselect
13235 Select frames to pass in output.
13237 This filter accepts the following options:
13242 Set expression, which is evaluated for each input frame.
13244 If the expression is evaluated to zero, the frame is discarded.
13246 If the evaluation result is negative or NaN, the frame is sent to the
13247 first output; otherwise it is sent to the output with index
13248 @code{ceil(val)-1}, assuming that the input index starts from 0.
13250 For example a value of @code{1.2} corresponds to the output with index
13251 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
13254 Set the number of outputs. The output to which to send the selected
13255 frame is based on the result of the evaluation. Default value is 1.
13258 The expression can contain the following constants:
13262 The (sequential) number of the filtered frame, starting from 0.
13265 The (sequential) number of the selected frame, starting from 0.
13267 @item prev_selected_n
13268 The sequential number of the last selected frame. It's NAN if undefined.
13271 The timebase of the input timestamps.
13274 The PTS (Presentation TimeStamp) of the filtered video frame,
13275 expressed in @var{TB} units. It's NAN if undefined.
13278 The PTS of the filtered video frame,
13279 expressed in seconds. It's NAN if undefined.
13282 The PTS of the previously filtered video frame. It's NAN if undefined.
13284 @item prev_selected_pts
13285 The PTS of the last previously filtered video frame. It's NAN if undefined.
13287 @item prev_selected_t
13288 The PTS of the last previously selected video frame. It's NAN if undefined.
13291 The PTS of the first video frame in the video. It's NAN if undefined.
13294 The time of the first video frame in the video. It's NAN if undefined.
13296 @item pict_type @emph{(video only)}
13297 The type of the filtered frame. It can assume one of the following
13309 @item interlace_type @emph{(video only)}
13310 The frame interlace type. It can assume one of the following values:
13313 The frame is progressive (not interlaced).
13315 The frame is top-field-first.
13317 The frame is bottom-field-first.
13320 @item consumed_sample_n @emph{(audio only)}
13321 the number of selected samples before the current frame
13323 @item samples_n @emph{(audio only)}
13324 the number of samples in the current frame
13326 @item sample_rate @emph{(audio only)}
13327 the input sample rate
13330 This is 1 if the filtered frame is a key-frame, 0 otherwise.
13333 the position in the file of the filtered frame, -1 if the information
13334 is not available (e.g. for synthetic video)
13336 @item scene @emph{(video only)}
13337 value between 0 and 1 to indicate a new scene; a low value reflects a low
13338 probability for the current frame to introduce a new scene, while a higher
13339 value means the current frame is more likely to be one (see the example below)
13341 @item concatdec_select
13342 The concat demuxer can select only part of a concat input file by setting an
13343 inpoint and an outpoint, but the output packets may not be entirely contained
13344 in the selected interval. By using this variable, it is possible to skip frames
13345 generated by the concat demuxer which are not exactly contained in the selected
13348 This works by comparing the frame pts against the @var{lavf.concat.start_time}
13349 and the @var{lavf.concat.duration} packet metadata values which are also
13350 present in the decoded frames.
13352 The @var{concatdec_select} variable is -1 if the frame pts is at least
13353 start_time and either the duration metadata is missing or the frame pts is less
13354 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
13357 That basically means that an input frame is selected if its pts is within the
13358 interval set by the concat demuxer.
13362 The default value of the select expression is "1".
13364 @subsection Examples
13368 Select all frames in input:
13373 The example above is the same as:
13385 Select only I-frames:
13387 select='eq(pict_type\,I)'
13391 Select one frame every 100:
13393 select='not(mod(n\,100))'
13397 Select only frames contained in the 10-20 time interval:
13399 select=between(t\,10\,20)
13403 Select only I frames contained in the 10-20 time interval:
13405 select=between(t\,10\,20)*eq(pict_type\,I)
13409 Select frames with a minimum distance of 10 seconds:
13411 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
13415 Use aselect to select only audio frames with samples number > 100:
13417 aselect='gt(samples_n\,100)'
13421 Create a mosaic of the first scenes:
13423 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
13426 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
13430 Send even and odd frames to separate outputs, and compose them:
13432 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
13436 Select useful frames from an ffconcat file which is using inpoints and
13437 outpoints but where the source files are not intra frame only.
13439 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
13443 @section selectivecolor
13445 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
13446 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
13447 by the "purity" of the color (that is, how saturated it already is).
13449 This filter is similar to the Adobe Photoshop Selective Color tool.
13451 The filter accepts the following options:
13454 @item correction_method
13455 Select color correction method.
13457 Available values are:
13460 Specified adjustments are applied "as-is" (added/subtracted to original pixel
13463 Specified adjustments are relative to the original component value.
13465 Default is @code{absolute}.
13467 Adjustments for red pixels (pixels where the red component is the maximum)
13469 Adjustments for yellow pixels (pixels where the blue component is the minimum)
13471 Adjustments for green pixels (pixels where the green component is the maximum)
13473 Adjustments for cyan pixels (pixels where the red component is the minimum)
13475 Adjustments for blue pixels (pixels where the blue component is the maximum)
13477 Adjustments for magenta pixels (pixels where the green component is the minimum)
13479 Adjustments for white pixels (pixels where all components are greater than 128)
13481 Adjustments for all pixels except pure black and pure white
13483 Adjustments for black pixels (pixels where all components are lesser than 128)
13485 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
13488 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
13489 4 space separated floating point adjustment values in the [-1,1] range,
13490 respectively to adjust the amount of cyan, magenta, yellow and black for the
13491 pixels of its range.
13493 @subsection Examples
13497 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
13498 increase magenta by 27% in blue areas:
13500 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
13504 Use a Photoshop selective color preset:
13506 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
13510 @section sendcmd, asendcmd
13512 Send commands to filters in the filtergraph.
13514 These filters read commands to be sent to other filters in the
13517 @code{sendcmd} must be inserted between two video filters,
13518 @code{asendcmd} must be inserted between two audio filters, but apart
13519 from that they act the same way.
13521 The specification of commands can be provided in the filter arguments
13522 with the @var{commands} option, or in a file specified by the
13523 @var{filename} option.
13525 These filters accept the following options:
13528 Set the commands to be read and sent to the other filters.
13530 Set the filename of the commands to be read and sent to the other
13534 @subsection Commands syntax
13536 A commands description consists of a sequence of interval
13537 specifications, comprising a list of commands to be executed when a
13538 particular event related to that interval occurs. The occurring event
13539 is typically the current frame time entering or leaving a given time
13542 An interval is specified by the following syntax:
13544 @var{START}[-@var{END}] @var{COMMANDS};
13547 The time interval is specified by the @var{START} and @var{END} times.
13548 @var{END} is optional and defaults to the maximum time.
13550 The current frame time is considered within the specified interval if
13551 it is included in the interval [@var{START}, @var{END}), that is when
13552 the time is greater or equal to @var{START} and is lesser than
13555 @var{COMMANDS} consists of a sequence of one or more command
13556 specifications, separated by ",", relating to that interval. The
13557 syntax of a command specification is given by:
13559 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
13562 @var{FLAGS} is optional and specifies the type of events relating to
13563 the time interval which enable sending the specified command, and must
13564 be a non-null sequence of identifier flags separated by "+" or "|" and
13565 enclosed between "[" and "]".
13567 The following flags are recognized:
13570 The command is sent when the current frame timestamp enters the
13571 specified interval. In other words, the command is sent when the
13572 previous frame timestamp was not in the given interval, and the
13576 The command is sent when the current frame timestamp leaves the
13577 specified interval. In other words, the command is sent when the
13578 previous frame timestamp was in the given interval, and the
13582 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
13585 @var{TARGET} specifies the target of the command, usually the name of
13586 the filter class or a specific filter instance name.
13588 @var{COMMAND} specifies the name of the command for the target filter.
13590 @var{ARG} is optional and specifies the optional list of argument for
13591 the given @var{COMMAND}.
13593 Between one interval specification and another, whitespaces, or
13594 sequences of characters starting with @code{#} until the end of line,
13595 are ignored and can be used to annotate comments.
13597 A simplified BNF description of the commands specification syntax
13600 @var{COMMAND_FLAG} ::= "enter" | "leave"
13601 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
13602 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
13603 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
13604 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
13605 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
13608 @subsection Examples
13612 Specify audio tempo change at second 4:
13614 asendcmd=c='4.0 atempo tempo 1.5',atempo
13618 Specify a list of drawtext and hue commands in a file.
13620 # show text in the interval 5-10
13621 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
13622 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
13624 # desaturate the image in the interval 15-20
13625 15.0-20.0 [enter] hue s 0,
13626 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
13628 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
13630 # apply an exponential saturation fade-out effect, starting from time 25
13631 25 [enter] hue s exp(25-t)
13634 A filtergraph allowing to read and process the above command list
13635 stored in a file @file{test.cmd}, can be specified with:
13637 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
13642 @section setpts, asetpts
13644 Change the PTS (presentation timestamp) of the input frames.
13646 @code{setpts} works on video frames, @code{asetpts} on audio frames.
13648 This filter accepts the following options:
13653 The expression which is evaluated for each frame to construct its timestamp.
13657 The expression is evaluated through the eval API and can contain the following
13662 frame rate, only defined for constant frame-rate video
13665 The presentation timestamp in input
13668 The count of the input frame for video or the number of consumed samples,
13669 not including the current frame for audio, starting from 0.
13671 @item NB_CONSUMED_SAMPLES
13672 The number of consumed samples, not including the current frame (only
13675 @item NB_SAMPLES, S
13676 The number of samples in the current frame (only audio)
13678 @item SAMPLE_RATE, SR
13679 The audio sample rate.
13682 The PTS of the first frame.
13685 the time in seconds of the first frame
13688 State whether the current frame is interlaced.
13691 the time in seconds of the current frame
13694 original position in the file of the frame, or undefined if undefined
13695 for the current frame
13698 The previous input PTS.
13701 previous input time in seconds
13704 The previous output PTS.
13707 previous output time in seconds
13710 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
13714 The wallclock (RTC) time at the start of the movie in microseconds.
13717 The timebase of the input timestamps.
13721 @subsection Examples
13725 Start counting PTS from zero
13727 setpts=PTS-STARTPTS
13731 Apply fast motion effect:
13737 Apply slow motion effect:
13743 Set fixed rate of 25 frames per second:
13749 Set fixed rate 25 fps with some jitter:
13751 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
13755 Apply an offset of 10 seconds to the input PTS:
13761 Generate timestamps from a "live source" and rebase onto the current timebase:
13763 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
13767 Generate timestamps by counting samples:
13774 @section settb, asettb
13776 Set the timebase to use for the output frames timestamps.
13777 It is mainly useful for testing timebase configuration.
13779 It accepts the following parameters:
13784 The expression which is evaluated into the output timebase.
13788 The value for @option{tb} is an arithmetic expression representing a
13789 rational. The expression can contain the constants "AVTB" (the default
13790 timebase), "intb" (the input timebase) and "sr" (the sample rate,
13791 audio only). Default value is "intb".
13793 @subsection Examples
13797 Set the timebase to 1/25:
13803 Set the timebase to 1/10:
13809 Set the timebase to 1001/1000:
13815 Set the timebase to 2*intb:
13821 Set the default timebase value:
13828 Convert input audio to a video output representing frequency spectrum
13829 logarithmically using Brown-Puckette constant Q transform algorithm with
13830 direct frequency domain coefficient calculation (but the transform itself
13831 is not really constant Q, instead the Q factor is actually variable/clamped),
13832 with musical tone scale, from E0 to D#10.
13834 The filter accepts the following options:
13838 Specify the video size for the output. It must be even. For the syntax of this option,
13839 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13840 Default value is @code{1920x1080}.
13843 Set the output frame rate. Default value is @code{25}.
13846 Set the bargraph height. It must be even. Default value is @code{-1} which
13847 computes the bargraph height automatically.
13850 Set the axis height. It must be even. Default value is @code{-1} which computes
13851 the axis height automatically.
13854 Set the sonogram height. It must be even. Default value is @code{-1} which
13855 computes the sonogram height automatically.
13858 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
13859 instead. Default value is @code{1}.
13861 @item sono_v, volume
13862 Specify the sonogram volume expression. It can contain variables:
13865 the @var{bar_v} evaluated expression
13866 @item frequency, freq, f
13867 the frequency where it is evaluated
13868 @item timeclamp, tc
13869 the value of @var{timeclamp} option
13873 @item a_weighting(f)
13874 A-weighting of equal loudness
13875 @item b_weighting(f)
13876 B-weighting of equal loudness
13877 @item c_weighting(f)
13878 C-weighting of equal loudness.
13880 Default value is @code{16}.
13882 @item bar_v, volume2
13883 Specify the bargraph volume expression. It can contain variables:
13886 the @var{sono_v} evaluated expression
13887 @item frequency, freq, f
13888 the frequency where it is evaluated
13889 @item timeclamp, tc
13890 the value of @var{timeclamp} option
13894 @item a_weighting(f)
13895 A-weighting of equal loudness
13896 @item b_weighting(f)
13897 B-weighting of equal loudness
13898 @item c_weighting(f)
13899 C-weighting of equal loudness.
13901 Default value is @code{sono_v}.
13903 @item sono_g, gamma
13904 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
13905 higher gamma makes the spectrum having more range. Default value is @code{3}.
13906 Acceptable range is @code{[1, 7]}.
13908 @item bar_g, gamma2
13909 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
13912 @item timeclamp, tc
13913 Specify the transform timeclamp. At low frequency, there is trade-off between
13914 accuracy in time domain and frequency domain. If timeclamp is lower,
13915 event in time domain is represented more accurately (such as fast bass drum),
13916 otherwise event in frequency domain is represented more accurately
13917 (such as bass guitar). Acceptable range is @code{[0.1, 1]}. Default value is @code{0.17}.
13920 Specify the transform base frequency. Default value is @code{20.01523126408007475},
13921 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
13924 Specify the transform end frequency. Default value is @code{20495.59681441799654},
13925 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
13928 This option is deprecated and ignored.
13931 Specify the transform length in time domain. Use this option to control accuracy
13932 trade-off between time domain and frequency domain at every frequency sample.
13933 It can contain variables:
13935 @item frequency, freq, f
13936 the frequency where it is evaluated
13937 @item timeclamp, tc
13938 the value of @var{timeclamp} option.
13940 Default value is @code{384*tc/(384+tc*f)}.
13943 Specify the transform count for every video frame. Default value is @code{6}.
13944 Acceptable range is @code{[1, 30]}.
13947 Specify the transform count for every single pixel. Default value is @code{0},
13948 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
13951 Specify font file for use with freetype to draw the axis. If not specified,
13952 use embedded font. Note that drawing with font file or embedded font is not
13953 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
13957 Specify font color expression. This is arithmetic expression that should return
13958 integer value 0xRRGGBB. It can contain variables:
13960 @item frequency, freq, f
13961 the frequency where it is evaluated
13962 @item timeclamp, tc
13963 the value of @var{timeclamp} option
13968 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
13969 @item r(x), g(x), b(x)
13970 red, green, and blue value of intensity x.
13972 Default value is @code{st(0, (midi(f)-59.5)/12);
13973 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
13974 r(1-ld(1)) + b(ld(1))}.
13977 Specify image file to draw the axis. This option override @var{fontfile} and
13978 @var{fontcolor} option.
13981 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
13982 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
13983 Default value is @code{1}.
13987 @subsection Examples
13991 Playing audio while showing the spectrum:
13993 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
13997 Same as above, but with frame rate 30 fps:
13999 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
14003 Playing at 1280x720:
14005 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
14009 Disable sonogram display:
14015 A1 and its harmonics: A1, A2, (near)E3, A3:
14017 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),
14018 asplit[a][out1]; [a] showcqt [out0]'
14022 Same as above, but with more accuracy in frequency domain:
14024 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),
14025 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
14031 bar_v=10:sono_v=bar_v*a_weighting(f)
14035 Custom gamma, now spectrum is linear to the amplitude.
14041 Custom tlength equation:
14043 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)))'
14047 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
14049 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
14053 Custom frequency range with custom axis using image file:
14055 axisfile=myaxis.png:basefreq=40:endfreq=10000
14061 Convert input audio to video output representing the audio power spectrum.
14062 Audio amplitude is on Y-axis while frequency is on X-axis.
14064 The filter accepts the following options:
14068 Specify size of video. For the syntax of this option, check the
14069 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14070 Default is @code{1024x512}.
14074 This set how each frequency bin will be represented.
14076 It accepts the following values:
14082 Default is @code{bar}.
14085 Set amplitude scale.
14087 It accepts the following values:
14101 Default is @code{log}.
14104 Set frequency scale.
14106 It accepts the following values:
14115 Reverse logarithmic scale.
14117 Default is @code{lin}.
14122 It accepts the following values:
14138 Default is @code{w2048}
14141 Set windowing function.
14143 It accepts the following values:
14160 Default is @code{hanning}.
14163 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
14164 which means optimal overlap for selected window function will be picked.
14167 Set time averaging. Setting this to 0 will display current maximal peaks.
14168 Default is @code{1}, which means time averaging is disabled.
14171 Specify list of colors separated by space or by '|' which will be used to
14172 draw channel frequencies. Unrecognized or missing colors will be replaced
14176 @section showspectrum
14178 Convert input audio to a video output, representing the audio frequency
14181 The filter accepts the following options:
14185 Specify the video size for the output. For the syntax of this option, check the
14186 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14187 Default value is @code{640x512}.
14190 Specify how the spectrum should slide along the window.
14192 It accepts the following values:
14195 the samples start again on the left when they reach the right
14197 the samples scroll from right to left
14199 frames are only produced when the samples reach the right
14202 Default value is @code{replace}.
14205 Specify display mode.
14207 It accepts the following values:
14210 all channels are displayed in the same row
14212 all channels are displayed in separate rows
14215 Default value is @samp{combined}.
14218 Specify display color mode.
14220 It accepts the following values:
14223 each channel is displayed in a separate color
14225 each channel is is displayed using the same color scheme
14228 Default value is @samp{channel}.
14231 Specify scale used for calculating intensity color values.
14233 It accepts the following values:
14238 square root, default
14245 Default value is @samp{sqrt}.
14248 Set saturation modifier for displayed colors. Negative values provide
14249 alternative color scheme. @code{0} is no saturation at all.
14250 Saturation must be in [-10.0, 10.0] range.
14251 Default value is @code{1}.
14254 Set window function.
14256 It accepts the following values:
14259 No samples pre-processing (do not expect this to be faster)
14268 Default value is @code{hann}.
14271 The usage is very similar to the showwaves filter; see the examples in that
14274 @subsection Examples
14278 Large window with logarithmic color scaling:
14280 showspectrum=s=1280x480:scale=log
14284 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
14286 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
14287 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
14291 @section showvolume
14293 Convert input audio volume to a video output.
14295 The filter accepts the following options:
14302 Set border width, allowed range is [0, 5]. Default is 1.
14305 Set channel width, allowed range is [80, 1080]. Default is 400.
14308 Set channel height, allowed range is [1, 100]. Default is 20.
14311 Set fade, allowed range is [0.001, 1]. Default is 0.95.
14314 Set volume color expression.
14316 The expression can use the following variables:
14320 Current max volume of channel in dB.
14323 Current channel number, starting from 0.
14327 If set, displays channel names. Default is enabled.
14330 If set, displays volume values. Default is enabled.
14335 Convert input audio to a video output, representing the samples waves.
14337 The filter accepts the following options:
14341 Specify the video size for the output. For the syntax of this option, check the
14342 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14343 Default value is @code{600x240}.
14348 Available values are:
14351 Draw a point for each sample.
14354 Draw a vertical line for each sample.
14357 Draw a point for each sample and a line between them.
14360 Draw a centered vertical line for each sample.
14363 Default value is @code{point}.
14366 Set the number of samples which are printed on the same column. A
14367 larger value will decrease the frame rate. Must be a positive
14368 integer. This option can be set only if the value for @var{rate}
14369 is not explicitly specified.
14372 Set the (approximate) output frame rate. This is done by setting the
14373 option @var{n}. Default value is "25".
14375 @item split_channels
14376 Set if channels should be drawn separately or overlap. Default value is 0.
14380 @subsection Examples
14384 Output the input file audio and the corresponding video representation
14387 amovie=a.mp3,asplit[out0],showwaves[out1]
14391 Create a synthetic signal and show it with showwaves, forcing a
14392 frame rate of 30 frames per second:
14394 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
14398 @section showwavespic
14400 Convert input audio to a single video frame, representing the samples waves.
14402 The filter accepts the following options:
14406 Specify the video size for the output. For the syntax of this option, check the
14407 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14408 Default value is @code{600x240}.
14410 @item split_channels
14411 Set if channels should be drawn separately or overlap. Default value is 0.
14414 @subsection Examples
14418 Extract a channel split representation of the wave form of a whole audio track
14419 in a 1024x800 picture using @command{ffmpeg}:
14421 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
14425 @section split, asplit
14427 Split input into several identical outputs.
14429 @code{asplit} works with audio input, @code{split} with video.
14431 The filter accepts a single parameter which specifies the number of outputs. If
14432 unspecified, it defaults to 2.
14434 @subsection Examples
14438 Create two separate outputs from the same input:
14440 [in] split [out0][out1]
14444 To create 3 or more outputs, you need to specify the number of
14447 [in] asplit=3 [out0][out1][out2]
14451 Create two separate outputs from the same input, one cropped and
14454 [in] split [splitout1][splitout2];
14455 [splitout1] crop=100:100:0:0 [cropout];
14456 [splitout2] pad=200:200:100:100 [padout];
14460 Create 5 copies of the input audio with @command{ffmpeg}:
14462 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
14468 Receive commands sent through a libzmq client, and forward them to
14469 filters in the filtergraph.
14471 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
14472 must be inserted between two video filters, @code{azmq} between two
14475 To enable these filters you need to install the libzmq library and
14476 headers and configure FFmpeg with @code{--enable-libzmq}.
14478 For more information about libzmq see:
14479 @url{http://www.zeromq.org/}
14481 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
14482 receives messages sent through a network interface defined by the
14483 @option{bind_address} option.
14485 The received message must be in the form:
14487 @var{TARGET} @var{COMMAND} [@var{ARG}]
14490 @var{TARGET} specifies the target of the command, usually the name of
14491 the filter class or a specific filter instance name.
14493 @var{COMMAND} specifies the name of the command for the target filter.
14495 @var{ARG} is optional and specifies the optional argument list for the
14496 given @var{COMMAND}.
14498 Upon reception, the message is processed and the corresponding command
14499 is injected into the filtergraph. Depending on the result, the filter
14500 will send a reply to the client, adopting the format:
14502 @var{ERROR_CODE} @var{ERROR_REASON}
14506 @var{MESSAGE} is optional.
14508 @subsection Examples
14510 Look at @file{tools/zmqsend} for an example of a zmq client which can
14511 be used to send commands processed by these filters.
14513 Consider the following filtergraph generated by @command{ffplay}
14515 ffplay -dumpgraph 1 -f lavfi "
14516 color=s=100x100:c=red [l];
14517 color=s=100x100:c=blue [r];
14518 nullsrc=s=200x100, zmq [bg];
14519 [bg][l] overlay [bg+l];
14520 [bg+l][r] overlay=x=100 "
14523 To change the color of the left side of the video, the following
14524 command can be used:
14526 echo Parsed_color_0 c yellow | tools/zmqsend
14529 To change the right side:
14531 echo Parsed_color_1 c pink | tools/zmqsend
14534 @c man end MULTIMEDIA FILTERS
14536 @chapter Multimedia Sources
14537 @c man begin MULTIMEDIA SOURCES
14539 Below is a description of the currently available multimedia sources.
14543 This is the same as @ref{movie} source, except it selects an audio
14549 Read audio and/or video stream(s) from a movie container.
14551 It accepts the following parameters:
14555 The name of the resource to read (not necessarily a file; it can also be a
14556 device or a stream accessed through some protocol).
14558 @item format_name, f
14559 Specifies the format assumed for the movie to read, and can be either
14560 the name of a container or an input device. If not specified, the
14561 format is guessed from @var{movie_name} or by probing.
14563 @item seek_point, sp
14564 Specifies the seek point in seconds. The frames will be output
14565 starting from this seek point. The parameter is evaluated with
14566 @code{av_strtod}, so the numerical value may be suffixed by an IS
14567 postfix. The default value is "0".
14570 Specifies the streams to read. Several streams can be specified,
14571 separated by "+". The source will then have as many outputs, in the
14572 same order. The syntax is explained in the ``Stream specifiers''
14573 section in the ffmpeg manual. Two special names, "dv" and "da" specify
14574 respectively the default (best suited) video and audio stream. Default
14575 is "dv", or "da" if the filter is called as "amovie".
14577 @item stream_index, si
14578 Specifies the index of the video stream to read. If the value is -1,
14579 the most suitable video stream will be automatically selected. The default
14580 value is "-1". Deprecated. If the filter is called "amovie", it will select
14581 audio instead of video.
14584 Specifies how many times to read the stream in sequence.
14585 If the value is less than 1, the stream will be read again and again.
14586 Default value is "1".
14588 Note that when the movie is looped the source timestamps are not
14589 changed, so it will generate non monotonically increasing timestamps.
14592 It allows overlaying a second video on top of the main input of
14593 a filtergraph, as shown in this graph:
14595 input -----------> deltapts0 --> overlay --> output
14598 movie --> scale--> deltapts1 -------+
14600 @subsection Examples
14604 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
14605 on top of the input labelled "in":
14607 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
14608 [in] setpts=PTS-STARTPTS [main];
14609 [main][over] overlay=16:16 [out]
14613 Read from a video4linux2 device, and overlay it on top of the input
14616 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
14617 [in] setpts=PTS-STARTPTS [main];
14618 [main][over] overlay=16:16 [out]
14622 Read the first video stream and the audio stream with id 0x81 from
14623 dvd.vob; the video is connected to the pad named "video" and the audio is
14624 connected to the pad named "audio":
14626 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
14630 @c man end MULTIMEDIA SOURCES