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 Apply cross fade from one input audio stream to another input audio stream.
324 The cross fade is applied for specified duration near the end of first stream.
326 The filter accepts the following options:
330 Specify the number of samples for which the cross fade effect has to last.
331 At the end of the cross fade effect the first input audio will be completely
332 silent. Default is 44100.
335 Specify the duration of the cross fade effect. See
336 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
337 for the accepted syntax.
338 By default the duration is determined by @var{nb_samples}.
339 If set this option is used instead of @var{nb_samples}.
342 Should first stream end overlap with second stream start. Default is enabled.
345 Set curve for cross fade transition for first stream.
348 Set curve for cross fade transition for second stream.
350 For description of available curve types see @ref{afade} filter description.
357 Cross fade from one input to another:
359 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
363 Cross fade from one input to another but without overlapping:
365 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
371 Delay one or more audio channels.
373 Samples in delayed channel are filled with silence.
375 The filter accepts the following option:
379 Set list of delays in milliseconds for each channel separated by '|'.
380 At least one delay greater than 0 should be provided.
381 Unused delays will be silently ignored. If number of given delays is
382 smaller than number of channels all remaining channels will not be delayed.
389 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
390 the second channel (and any other channels that may be present) unchanged.
398 Apply echoing to the input audio.
400 Echoes are reflected sound and can occur naturally amongst mountains
401 (and sometimes large buildings) when talking or shouting; digital echo
402 effects emulate this behaviour and are often used to help fill out the
403 sound of a single instrument or vocal. The time difference between the
404 original signal and the reflection is the @code{delay}, and the
405 loudness of the reflected signal is the @code{decay}.
406 Multiple echoes can have different delays and decays.
408 A description of the accepted parameters follows.
412 Set input gain of reflected signal. Default is @code{0.6}.
415 Set output gain of reflected signal. Default is @code{0.3}.
418 Set list of time intervals in milliseconds between original signal and reflections
419 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
420 Default is @code{1000}.
423 Set list of loudnesses of reflected signals separated by '|'.
424 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
425 Default is @code{0.5}.
432 Make it sound as if there are twice as many instruments as are actually playing:
434 aecho=0.8:0.88:60:0.4
438 If delay is very short, then it sound like a (metallic) robot playing music:
444 A longer delay will sound like an open air concert in the mountains:
446 aecho=0.8:0.9:1000:0.3
450 Same as above but with one more mountain:
452 aecho=0.8:0.9:1000|1800:0.3|0.25
458 Modify an audio signal according to the specified expressions.
460 This filter accepts one or more expressions (one for each channel),
461 which are evaluated and used to modify a corresponding audio signal.
463 It accepts the following parameters:
467 Set the '|'-separated expressions list for each separate channel. If
468 the number of input channels is greater than the number of
469 expressions, the last specified expression is used for the remaining
472 @item channel_layout, c
473 Set output channel layout. If not specified, the channel layout is
474 specified by the number of expressions. If set to @samp{same}, it will
475 use by default the same input channel layout.
478 Each expression in @var{exprs} can contain the following constants and functions:
482 channel number of the current expression
485 number of the evaluated sample, starting from 0
491 time of the evaluated sample expressed in seconds
494 @item nb_out_channels
495 input and output number of channels
498 the value of input channel with number @var{CH}
501 Note: this filter is slow. For faster processing you should use a
510 aeval=val(ch)/2:c=same
514 Invert phase of the second channel:
523 Apply fade-in/out effect to input audio.
525 A description of the accepted parameters follows.
529 Specify the effect type, can be either @code{in} for fade-in, or
530 @code{out} for a fade-out effect. Default is @code{in}.
532 @item start_sample, ss
533 Specify the number of the start sample for starting to apply the fade
534 effect. Default is 0.
537 Specify the number of samples for which the fade effect has to last. At
538 the end of the fade-in effect the output audio will have the same
539 volume as the input audio, at the end of the fade-out transition
540 the output audio will be silence. Default is 44100.
543 Specify the start time of the fade effect. Default is 0.
544 The value must be specified as a time duration; see
545 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
546 for the accepted syntax.
547 If set this option is used instead of @var{start_sample}.
550 Specify the duration of the fade effect. See
551 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
552 for the accepted syntax.
553 At the end of the fade-in effect the output audio will have the same
554 volume as the input audio, at the end of the fade-out transition
555 the output audio will be silence.
556 By default the duration is determined by @var{nb_samples}.
557 If set this option is used instead of @var{nb_samples}.
560 Set curve for fade transition.
562 It accepts the following values:
565 select triangular, linear slope (default)
567 select quarter of sine wave
569 select half of sine wave
571 select exponential sine wave
575 select inverted parabola
589 select inverted quarter of sine wave
591 select inverted half of sine wave
593 select double-exponential seat
595 select double-exponential sigmoid
603 Fade in first 15 seconds of audio:
609 Fade out last 25 seconds of a 900 seconds audio:
611 afade=t=out:st=875:d=25
618 Set output format constraints for the input audio. The framework will
619 negotiate the most appropriate format to minimize conversions.
621 It accepts the following parameters:
625 A '|'-separated list of requested sample formats.
628 A '|'-separated list of requested sample rates.
630 @item channel_layouts
631 A '|'-separated list of requested channel layouts.
633 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
634 for the required syntax.
637 If a parameter is omitted, all values are allowed.
639 Force the output to either unsigned 8-bit or signed 16-bit stereo
641 aformat=sample_fmts=u8|s16:channel_layouts=stereo
646 A gate is mainly used to reduce lower parts of a signal. This kind of signal
647 processing reduces disturbing noise between useful signals.
649 Gating is done by detecting the volume below a chosen level @var{threshold}
650 and divide it by the factor set with @var{ratio}. The bottom of the noise
651 floor is set via @var{range}. Because an exact manipulation of the signal
652 would cause distortion of the waveform the reduction can be levelled over
653 time. This is done by setting @var{attack} and @var{release}.
655 @var{attack} determines how long the signal has to fall below the threshold
656 before any reduction will occur and @var{release} sets the time the signal
657 has to raise above the threshold to reduce the reduction again.
658 Shorter signals than the chosen attack time will be left untouched.
662 Set input level before filtering.
665 Set the level of gain reduction when the signal is below the threshold.
668 If a signal rises above this level the gain reduction is released.
671 Set a ratio about which the signal is reduced.
674 Amount of milliseconds the signal has to rise above the threshold before gain
678 Amount of milliseconds the signal has to fall below the threshold before the
679 reduction is increased again.
682 Set amount of amplification of signal after processing.
685 Curve the sharp knee around the threshold to enter gain reduction more softly.
688 Choose if exact signal should be taken for detection or an RMS like one.
691 Choose if the average level between all channels or the louder channel affects
697 The limiter prevents input signal from raising over a desired threshold.
698 This limiter uses lookahead technology to prevent your signal from distorting.
699 It means that there is a small delay after signal is processed. Keep in mind
700 that the delay it produces is the attack time you set.
702 The filter accepts the following options:
706 Don't let signals above this level pass the limiter. The removed amplitude is
707 added automatically. Default is 1.
710 The limiter will reach its attenuation level in this amount of time in
711 milliseconds. Default is 5 milliseconds.
714 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
715 Default is 50 milliseconds.
718 When gain reduction is always needed ASC takes care of releasing to an
719 average reduction level rather than reaching a reduction of 0 in the release
723 Select how much the release time is affected by ASC, 0 means nearly no changes
724 in release time while 1 produces higher release times.
727 Depending on picked setting it is recommended to upsample input 2x or 4x times
728 with @ref{aresample} before applying this filter.
732 Apply a two-pole all-pass filter with central frequency (in Hz)
733 @var{frequency}, and filter-width @var{width}.
734 An all-pass filter changes the audio's frequency to phase relationship
735 without changing its frequency to amplitude relationship.
737 The filter accepts the following options:
744 Set method to specify band-width of filter.
757 Specify the band-width of a filter in width_type units.
763 Merge two or more audio streams into a single multi-channel stream.
765 The filter accepts the following options:
770 Set the number of inputs. Default is 2.
774 If the channel layouts of the inputs are disjoint, and therefore compatible,
775 the channel layout of the output will be set accordingly and the channels
776 will be reordered as necessary. If the channel layouts of the inputs are not
777 disjoint, the output will have all the channels of the first input then all
778 the channels of the second input, in that order, and the channel layout of
779 the output will be the default value corresponding to the total number of
782 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
783 is FC+BL+BR, then the output will be in 5.1, with the channels in the
784 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
785 first input, b1 is the first channel of the second input).
787 On the other hand, if both input are in stereo, the output channels will be
788 in the default order: a1, a2, b1, b2, and the channel layout will be
789 arbitrarily set to 4.0, which may or may not be the expected value.
791 All inputs must have the same sample rate, and format.
793 If inputs do not have the same duration, the output will stop with the
800 Merge two mono files into a stereo stream:
802 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
806 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
808 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
814 Mixes multiple audio inputs into a single output.
816 Note that this filter only supports float samples (the @var{amerge}
817 and @var{pan} audio filters support many formats). If the @var{amix}
818 input has integer samples then @ref{aresample} will be automatically
819 inserted to perform the conversion to float samples.
823 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
825 will mix 3 input audio streams to a single output with the same duration as the
826 first input and a dropout transition time of 3 seconds.
828 It accepts the following parameters:
832 The number of inputs. If unspecified, it defaults to 2.
835 How to determine the end-of-stream.
839 The duration of the longest input. (default)
842 The duration of the shortest input.
845 The duration of the first input.
849 @item dropout_transition
850 The transition time, in seconds, for volume renormalization when an input
851 stream ends. The default value is 2 seconds.
857 Pass the audio source unchanged to the output.
861 Pad the end of an audio stream with silence.
863 This can be used together with @command{ffmpeg} @option{-shortest} to
864 extend audio streams to the same length as the video stream.
866 A description of the accepted options follows.
870 Set silence packet size. Default value is 4096.
873 Set the number of samples of silence to add to the end. After the
874 value is reached, the stream is terminated. This option is mutually
875 exclusive with @option{whole_len}.
878 Set the minimum total number of samples in the output audio stream. If
879 the value is longer than the input audio length, silence is added to
880 the end, until the value is reached. This option is mutually exclusive
881 with @option{pad_len}.
884 If neither the @option{pad_len} nor the @option{whole_len} option is
885 set, the filter will add silence to the end of the input stream
892 Add 1024 samples of silence to the end of the input:
898 Make sure the audio output will contain at least 10000 samples, pad
899 the input with silence if required:
905 Use @command{ffmpeg} to pad the audio input with silence, so that the
906 video stream will always result the shortest and will be converted
907 until the end in the output file when using the @option{shortest}
910 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
915 Add a phasing effect to the input audio.
917 A phaser filter creates series of peaks and troughs in the frequency spectrum.
918 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
920 A description of the accepted parameters follows.
924 Set input gain. Default is 0.4.
927 Set output gain. Default is 0.74
930 Set delay in milliseconds. Default is 3.0.
933 Set decay. Default is 0.4.
936 Set modulation speed in Hz. Default is 0.5.
939 Set modulation type. Default is triangular.
941 It accepts the following values:
951 Resample the input audio to the specified parameters, using the
952 libswresample library. If none are specified then the filter will
953 automatically convert between its input and output.
955 This filter is also able to stretch/squeeze the audio data to make it match
956 the timestamps or to inject silence / cut out audio to make it match the
957 timestamps, do a combination of both or do neither.
959 The filter accepts the syntax
960 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
961 expresses a sample rate and @var{resampler_options} is a list of
962 @var{key}=@var{value} pairs, separated by ":". See the
963 ffmpeg-resampler manual for the complete list of supported options.
969 Resample the input audio to 44100Hz:
975 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
976 samples per second compensation:
982 @section asetnsamples
984 Set the number of samples per each output audio frame.
986 The last output packet may contain a different number of samples, as
987 the filter will flush all the remaining samples when the input audio
990 The filter accepts the following options:
994 @item nb_out_samples, n
995 Set the number of frames per each output audio frame. The number is
996 intended as the number of samples @emph{per each channel}.
997 Default value is 1024.
1000 If set to 1, the filter will pad the last audio frame with zeroes, so
1001 that the last frame will contain the same number of samples as the
1002 previous ones. Default value is 1.
1005 For example, to set the number of per-frame samples to 1234 and
1006 disable padding for the last frame, use:
1008 asetnsamples=n=1234:p=0
1013 Set the sample rate without altering the PCM data.
1014 This will result in a change of speed and pitch.
1016 The filter accepts the following options:
1019 @item sample_rate, r
1020 Set the output sample rate. Default is 44100 Hz.
1025 Show a line containing various information for each input audio frame.
1026 The input audio is not modified.
1028 The shown line contains a sequence of key/value pairs of the form
1029 @var{key}:@var{value}.
1031 The following values are shown in the output:
1035 The (sequential) number of the input frame, starting from 0.
1038 The presentation timestamp of the input frame, in time base units; the time base
1039 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1042 The presentation timestamp of the input frame in seconds.
1045 position of the frame in the input stream, -1 if this information in
1046 unavailable and/or meaningless (for example in case of synthetic audio)
1055 The sample rate for the audio frame.
1058 The number of samples (per channel) in the frame.
1061 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1062 audio, the data is treated as if all the planes were concatenated.
1064 @item plane_checksums
1065 A list of Adler-32 checksums for each data plane.
1071 Display time domain statistical information about the audio channels.
1072 Statistics are calculated and displayed for each audio channel and,
1073 where applicable, an overall figure is also given.
1075 It accepts the following option:
1078 Short window length in seconds, used for peak and trough RMS measurement.
1079 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1083 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1084 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1087 Available keys for each channel are:
1118 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1119 this @code{lavfi.astats.Overall.Peak_count}.
1121 For description what each key means read below.
1124 Set number of frame after which stats are going to be recalculated.
1125 Default is disabled.
1128 A description of each shown parameter follows:
1132 Mean amplitude displacement from zero.
1135 Minimal sample level.
1138 Maximal sample level.
1140 @item Min difference
1141 Minimal difference between two consecutive samples.
1143 @item Max difference
1144 Maximal difference between two consecutive samples.
1146 @item Mean difference
1147 Mean difference between two consecutive samples.
1148 The average of each difference between two consecutive samples.
1152 Standard peak and RMS level measured in dBFS.
1156 Peak and trough values for RMS level measured over a short window.
1159 Standard ratio of peak to RMS level (note: not in dB).
1162 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1163 (i.e. either @var{Min level} or @var{Max level}).
1166 Number of occasions (not the number of samples) that the signal attained either
1167 @var{Min level} or @var{Max level}.
1170 Overall bit depth of audio. Number of bits used for each sample.
1173 @section astreamsync
1175 Forward two audio streams and control the order the buffers are forwarded.
1177 The filter accepts the following options:
1181 Set the expression deciding which stream should be
1182 forwarded next: if the result is negative, the first stream is forwarded; if
1183 the result is positive or zero, the second stream is forwarded. It can use
1184 the following variables:
1188 number of buffers forwarded so far on each stream
1190 number of samples forwarded so far on each stream
1192 current timestamp of each stream
1195 The default value is @code{t1-t2}, which means to always forward the stream
1196 that has a smaller timestamp.
1199 @subsection Examples
1201 Stress-test @code{amerge} by randomly sending buffers on the wrong
1202 input, while avoiding too much of a desynchronization:
1204 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
1205 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
1211 Synchronize audio data with timestamps by squeezing/stretching it and/or
1212 dropping samples/adding silence when needed.
1214 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1216 It accepts the following parameters:
1220 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1221 by default. When disabled, time gaps are covered with silence.
1224 The minimum difference between timestamps and audio data (in seconds) to trigger
1225 adding/dropping samples. The default value is 0.1. If you get an imperfect
1226 sync with this filter, try setting this parameter to 0.
1229 The maximum compensation in samples per second. Only relevant with compensate=1.
1230 The default value is 500.
1233 Assume that the first PTS should be this value. The time base is 1 / sample
1234 rate. This allows for padding/trimming at the start of the stream. By default,
1235 no assumption is made about the first frame's expected PTS, so no padding or
1236 trimming is done. For example, this could be set to 0 to pad the beginning with
1237 silence if an audio stream starts after the video stream or to trim any samples
1238 with a negative PTS due to encoder delay.
1246 The filter accepts exactly one parameter, the audio tempo. If not
1247 specified then the filter will assume nominal 1.0 tempo. Tempo must
1248 be in the [0.5, 2.0] range.
1250 @subsection Examples
1254 Slow down audio to 80% tempo:
1260 To speed up audio to 125% tempo:
1268 Trim the input so that the output contains one continuous subpart of the input.
1270 It accepts the following parameters:
1273 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1274 sample with the timestamp @var{start} will be the first sample in the output.
1277 Specify time of the first audio sample that will be dropped, i.e. the
1278 audio sample immediately preceding the one with the timestamp @var{end} will be
1279 the last sample in the output.
1282 Same as @var{start}, except this option sets the start timestamp in samples
1286 Same as @var{end}, except this option sets the end timestamp in samples instead
1290 The maximum duration of the output in seconds.
1293 The number of the first sample that should be output.
1296 The number of the first sample that should be dropped.
1299 @option{start}, @option{end}, and @option{duration} are expressed as time
1300 duration specifications; see
1301 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1303 Note that the first two sets of the start/end options and the @option{duration}
1304 option look at the frame timestamp, while the _sample options simply count the
1305 samples that pass through the filter. So start/end_pts and start/end_sample will
1306 give different results when the timestamps are wrong, inexact or do not start at
1307 zero. Also note that this filter does not modify the timestamps. If you wish
1308 to have the output timestamps start at zero, insert the asetpts filter after the
1311 If multiple start or end options are set, this filter tries to be greedy and
1312 keep all samples that match at least one of the specified constraints. To keep
1313 only the part that matches all the constraints at once, chain multiple atrim
1316 The defaults are such that all the input is kept. So it is possible to set e.g.
1317 just the end values to keep everything before the specified time.
1322 Drop everything except the second minute of input:
1324 ffmpeg -i INPUT -af atrim=60:120
1328 Keep only the first 1000 samples:
1330 ffmpeg -i INPUT -af atrim=end_sample=1000
1337 Apply a two-pole Butterworth band-pass filter with central
1338 frequency @var{frequency}, and (3dB-point) band-width width.
1339 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1340 instead of the default: constant 0dB peak gain.
1341 The filter roll off at 6dB per octave (20dB per decade).
1343 The filter accepts the following options:
1347 Set the filter's central frequency. Default is @code{3000}.
1350 Constant skirt gain if set to 1. Defaults to 0.
1353 Set method to specify band-width of filter.
1366 Specify the band-width of a filter in width_type units.
1371 Apply a two-pole Butterworth band-reject filter with central
1372 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1373 The filter roll off at 6dB per octave (20dB per decade).
1375 The filter accepts the following options:
1379 Set the filter's central frequency. Default is @code{3000}.
1382 Set method to specify band-width of filter.
1395 Specify the band-width of a filter in width_type units.
1400 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1401 shelving filter with a response similar to that of a standard
1402 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1404 The filter accepts the following options:
1408 Give the gain at 0 Hz. Its useful range is about -20
1409 (for a large cut) to +20 (for a large boost).
1410 Beware of clipping when using a positive gain.
1413 Set the filter's central frequency and so can be used
1414 to extend or reduce the frequency range to be boosted or cut.
1415 The default value is @code{100} Hz.
1418 Set method to specify band-width of filter.
1431 Determine how steep is the filter's shelf transition.
1436 Apply a biquad IIR filter with the given coefficients.
1437 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1438 are the numerator and denominator coefficients respectively.
1441 Bauer stereo to binaural transformation, which improves headphone listening of
1442 stereo audio records.
1444 It accepts the following parameters:
1448 Pre-defined crossfeed level.
1452 Default level (fcut=700, feed=50).
1455 Chu Moy circuit (fcut=700, feed=60).
1458 Jan Meier circuit (fcut=650, feed=95).
1463 Cut frequency (in Hz).
1472 Remap input channels to new locations.
1474 It accepts the following parameters:
1476 @item channel_layout
1477 The channel layout of the output stream.
1480 Map channels from input to output. The argument is a '|'-separated list of
1481 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1482 @var{in_channel} form. @var{in_channel} can be either the name of the input
1483 channel (e.g. FL for front left) or its index in the input channel layout.
1484 @var{out_channel} is the name of the output channel or its index in the output
1485 channel layout. If @var{out_channel} is not given then it is implicitly an
1486 index, starting with zero and increasing by one for each mapping.
1489 If no mapping is present, the filter will implicitly map input channels to
1490 output channels, preserving indices.
1492 For example, assuming a 5.1+downmix input MOV file,
1494 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1496 will create an output WAV file tagged as stereo from the downmix channels of
1499 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1501 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1504 @section channelsplit
1506 Split each channel from an input audio stream into a separate output stream.
1508 It accepts the following parameters:
1510 @item channel_layout
1511 The channel layout of the input stream. The default is "stereo".
1514 For example, assuming a stereo input MP3 file,
1516 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1518 will create an output Matroska file with two audio streams, one containing only
1519 the left channel and the other the right channel.
1521 Split a 5.1 WAV file into per-channel files:
1523 ffmpeg -i in.wav -filter_complex
1524 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1525 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1526 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1531 Add a chorus effect to the audio.
1533 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1535 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1536 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1537 The modulation depth defines the range the modulated delay is played before or after
1538 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1539 sound tuned around the original one, like in a chorus where some vocals are slightly
1542 It accepts the following parameters:
1545 Set input gain. Default is 0.4.
1548 Set output gain. Default is 0.4.
1551 Set delays. A typical delay is around 40ms to 60ms.
1563 @subsection Examples
1569 chorus=0.7:0.9:55:0.4:0.25:2
1575 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1579 Fuller sounding chorus with three delays:
1581 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
1586 Compress or expand the audio's dynamic range.
1588 It accepts the following parameters:
1594 A list of times in seconds for each channel over which the instantaneous level
1595 of the input signal is averaged to determine its volume. @var{attacks} refers to
1596 increase of volume and @var{decays} refers to decrease of volume. For most
1597 situations, the attack time (response to the audio getting louder) should be
1598 shorter than the decay time, because the human ear is more sensitive to sudden
1599 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1600 a typical value for decay is 0.8 seconds.
1601 If specified number of attacks & decays is lower than number of channels, the last
1602 set attack/decay will be used for all remaining channels.
1605 A list of points for the transfer function, specified in dB relative to the
1606 maximum possible signal amplitude. Each key points list must be defined using
1607 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1608 @code{x0/y0 x1/y1 x2/y2 ....}
1610 The input values must be in strictly increasing order but the transfer function
1611 does not have to be monotonically rising. The point @code{0/0} is assumed but
1612 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1613 function are @code{-70/-70|-60/-20}.
1616 Set the curve radius in dB for all joints. It defaults to 0.01.
1619 Set the additional gain in dB to be applied at all points on the transfer
1620 function. This allows for easy adjustment of the overall gain.
1624 Set an initial volume, in dB, to be assumed for each channel when filtering
1625 starts. This permits the user to supply a nominal level initially, so that, for
1626 example, a very large gain is not applied to initial signal levels before the
1627 companding has begun to operate. A typical value for audio which is initially
1628 quiet is -90 dB. It defaults to 0.
1631 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1632 delayed before being fed to the volume adjuster. Specifying a delay
1633 approximately equal to the attack/decay times allows the filter to effectively
1634 operate in predictive rather than reactive mode. It defaults to 0.
1638 @subsection Examples
1642 Make music with both quiet and loud passages suitable for listening to in a
1645 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1648 Another example for audio with whisper and explosion parts:
1650 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
1654 A noise gate for when the noise is at a lower level than the signal:
1656 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1660 Here is another noise gate, this time for when the noise is at a higher level
1661 than the signal (making it, in some ways, similar to squelch):
1663 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1668 Apply a DC shift to the audio.
1670 This can be useful to remove a DC offset (caused perhaps by a hardware problem
1671 in the recording chain) from the audio. The effect of a DC offset is reduced
1672 headroom and hence volume. The @ref{astats} filter can be used to determine if
1673 a signal has a DC offset.
1677 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
1681 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
1682 used to prevent clipping.
1686 Dynamic Audio Normalizer.
1688 This filter applies a certain amount of gain to the input audio in order
1689 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
1690 contrast to more "simple" normalization algorithms, the Dynamic Audio
1691 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
1692 This allows for applying extra gain to the "quiet" sections of the audio
1693 while avoiding distortions or clipping the "loud" sections. In other words:
1694 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
1695 sections, in the sense that the volume of each section is brought to the
1696 same target level. Note, however, that the Dynamic Audio Normalizer achieves
1697 this goal *without* applying "dynamic range compressing". It will retain 100%
1698 of the dynamic range *within* each section of the audio file.
1702 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
1703 Default is 500 milliseconds.
1704 The Dynamic Audio Normalizer processes the input audio in small chunks,
1705 referred to as frames. This is required, because a peak magnitude has no
1706 meaning for just a single sample value. Instead, we need to determine the
1707 peak magnitude for a contiguous sequence of sample values. While a "standard"
1708 normalizer would simply use the peak magnitude of the complete file, the
1709 Dynamic Audio Normalizer determines the peak magnitude individually for each
1710 frame. The length of a frame is specified in milliseconds. By default, the
1711 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
1712 been found to give good results with most files.
1713 Note that the exact frame length, in number of samples, will be determined
1714 automatically, based on the sampling rate of the individual input audio file.
1717 Set the Gaussian filter window size. In range from 3 to 301, must be odd
1718 number. Default is 31.
1719 Probably the most important parameter of the Dynamic Audio Normalizer is the
1720 @code{window size} of the Gaussian smoothing filter. The filter's window size
1721 is specified in frames, centered around the current frame. For the sake of
1722 simplicity, this must be an odd number. Consequently, the default value of 31
1723 takes into account the current frame, as well as the 15 preceding frames and
1724 the 15 subsequent frames. Using a larger window results in a stronger
1725 smoothing effect and thus in less gain variation, i.e. slower gain
1726 adaptation. Conversely, using a smaller window results in a weaker smoothing
1727 effect and thus in more gain variation, i.e. faster gain adaptation.
1728 In other words, the more you increase this value, the more the Dynamic Audio
1729 Normalizer will behave like a "traditional" normalization filter. On the
1730 contrary, the more you decrease this value, the more the Dynamic Audio
1731 Normalizer will behave like a dynamic range compressor.
1734 Set the target peak value. This specifies the highest permissible magnitude
1735 level for the normalized audio input. This filter will try to approach the
1736 target peak magnitude as closely as possible, but at the same time it also
1737 makes sure that the normalized signal will never exceed the peak magnitude.
1738 A frame's maximum local gain factor is imposed directly by the target peak
1739 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
1740 It is not recommended to go above this value.
1743 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
1744 The Dynamic Audio Normalizer determines the maximum possible (local) gain
1745 factor for each input frame, i.e. the maximum gain factor that does not
1746 result in clipping or distortion. The maximum gain factor is determined by
1747 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
1748 additionally bounds the frame's maximum gain factor by a predetermined
1749 (global) maximum gain factor. This is done in order to avoid excessive gain
1750 factors in "silent" or almost silent frames. By default, the maximum gain
1751 factor is 10.0, For most inputs the default value should be sufficient and
1752 it usually is not recommended to increase this value. Though, for input
1753 with an extremely low overall volume level, it may be necessary to allow even
1754 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
1755 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
1756 Instead, a "sigmoid" threshold function will be applied. This way, the
1757 gain factors will smoothly approach the threshold value, but never exceed that
1761 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
1762 By default, the Dynamic Audio Normalizer performs "peak" normalization.
1763 This means that the maximum local gain factor for each frame is defined
1764 (only) by the frame's highest magnitude sample. This way, the samples can
1765 be amplified as much as possible without exceeding the maximum signal
1766 level, i.e. without clipping. Optionally, however, the Dynamic Audio
1767 Normalizer can also take into account the frame's root mean square,
1768 abbreviated RMS. In electrical engineering, the RMS is commonly used to
1769 determine the power of a time-varying signal. It is therefore considered
1770 that the RMS is a better approximation of the "perceived loudness" than
1771 just looking at the signal's peak magnitude. Consequently, by adjusting all
1772 frames to a constant RMS value, a uniform "perceived loudness" can be
1773 established. If a target RMS value has been specified, a frame's local gain
1774 factor is defined as the factor that would result in exactly that RMS value.
1775 Note, however, that the maximum local gain factor is still restricted by the
1776 frame's highest magnitude sample, in order to prevent clipping.
1779 Enable channels coupling. By default is enabled.
1780 By default, the Dynamic Audio Normalizer will amplify all channels by the same
1781 amount. This means the same gain factor will be applied to all channels, i.e.
1782 the maximum possible gain factor is determined by the "loudest" channel.
1783 However, in some recordings, it may happen that the volume of the different
1784 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
1785 In this case, this option can be used to disable the channel coupling. This way,
1786 the gain factor will be determined independently for each channel, depending
1787 only on the individual channel's highest magnitude sample. This allows for
1788 harmonizing the volume of the different channels.
1791 Enable DC bias correction. By default is disabled.
1792 An audio signal (in the time domain) is a sequence of sample values.
1793 In the Dynamic Audio Normalizer these sample values are represented in the
1794 -1.0 to 1.0 range, regardless of the original input format. Normally, the
1795 audio signal, or "waveform", should be centered around the zero point.
1796 That means if we calculate the mean value of all samples in a file, or in a
1797 single frame, then the result should be 0.0 or at least very close to that
1798 value. If, however, there is a significant deviation of the mean value from
1799 0.0, in either positive or negative direction, this is referred to as a
1800 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
1801 Audio Normalizer provides optional DC bias correction.
1802 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
1803 the mean value, or "DC correction" offset, of each input frame and subtract
1804 that value from all of the frame's sample values which ensures those samples
1805 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
1806 boundaries, the DC correction offset values will be interpolated smoothly
1807 between neighbouring frames.
1810 Enable alternative boundary mode. By default is disabled.
1811 The Dynamic Audio Normalizer takes into account a certain neighbourhood
1812 around each frame. This includes the preceding frames as well as the
1813 subsequent frames. However, for the "boundary" frames, located at the very
1814 beginning and at the very end of the audio file, not all neighbouring
1815 frames are available. In particular, for the first few frames in the audio
1816 file, the preceding frames are not known. And, similarly, for the last few
1817 frames in the audio file, the subsequent frames are not known. Thus, the
1818 question arises which gain factors should be assumed for the missing frames
1819 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
1820 to deal with this situation. The default boundary mode assumes a gain factor
1821 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
1822 "fade out" at the beginning and at the end of the input, respectively.
1825 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
1826 By default, the Dynamic Audio Normalizer does not apply "traditional"
1827 compression. This means that signal peaks will not be pruned and thus the
1828 full dynamic range will be retained within each local neighbourhood. However,
1829 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
1830 normalization algorithm with a more "traditional" compression.
1831 For this purpose, the Dynamic Audio Normalizer provides an optional compression
1832 (thresholding) function. If (and only if) the compression feature is enabled,
1833 all input frames will be processed by a soft knee thresholding function prior
1834 to the actual normalization process. Put simply, the thresholding function is
1835 going to prune all samples whose magnitude exceeds a certain threshold value.
1836 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
1837 value. Instead, the threshold value will be adjusted for each individual
1839 In general, smaller parameters result in stronger compression, and vice versa.
1840 Values below 3.0 are not recommended, because audible distortion may appear.
1845 Make audio easier to listen to on headphones.
1847 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1848 so that when listened to on headphones the stereo image is moved from
1849 inside your head (standard for headphones) to outside and in front of
1850 the listener (standard for speakers).
1856 Apply a two-pole peaking equalisation (EQ) filter. With this
1857 filter, the signal-level at and around a selected frequency can
1858 be increased or decreased, whilst (unlike bandpass and bandreject
1859 filters) that at all other frequencies is unchanged.
1861 In order to produce complex equalisation curves, this filter can
1862 be given several times, each with a different central frequency.
1864 The filter accepts the following options:
1868 Set the filter's central frequency in Hz.
1871 Set method to specify band-width of filter.
1884 Specify the band-width of a filter in width_type units.
1887 Set the required gain or attenuation in dB.
1888 Beware of clipping when using a positive gain.
1891 @subsection Examples
1894 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1896 equalizer=f=1000:width_type=h:width=200:g=-10
1900 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1902 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
1906 @section extrastereo
1908 Linearly increases the difference between left and right channels which
1909 adds some sort of "live" effect to playback.
1911 The filter accepts the following option:
1915 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
1916 (average of both channels), with 1.0 sound will be unchanged, with
1917 -1.0 left and right channels will be swapped.
1920 Enable clipping. By default is enabled.
1924 Apply a flanging effect to the audio.
1926 The filter accepts the following options:
1930 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
1933 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
1936 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
1940 Set percentage of delayed signal mixed with original. Range from 0 to 100.
1941 Default value is 71.
1944 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
1947 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
1948 Default value is @var{sinusoidal}.
1951 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
1952 Default value is 25.
1955 Set delay-line interpolation, @var{linear} or @var{quadratic}.
1956 Default is @var{linear}.
1961 Apply a high-pass filter with 3dB point frequency.
1962 The filter can be either single-pole, or double-pole (the default).
1963 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1965 The filter accepts the following options:
1969 Set frequency in Hz. Default is 3000.
1972 Set number of poles. Default is 2.
1975 Set method to specify band-width of filter.
1988 Specify the band-width of a filter in width_type units.
1989 Applies only to double-pole filter.
1990 The default is 0.707q and gives a Butterworth response.
1995 Join multiple input streams into one multi-channel stream.
1997 It accepts the following parameters:
2001 The number of input streams. It defaults to 2.
2003 @item channel_layout
2004 The desired output channel layout. It defaults to stereo.
2007 Map channels from inputs to output. The argument is a '|'-separated list of
2008 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2009 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2010 can be either the name of the input channel (e.g. FL for front left) or its
2011 index in the specified input stream. @var{out_channel} is the name of the output
2015 The filter will attempt to guess the mappings when they are not specified
2016 explicitly. It does so by first trying to find an unused matching input channel
2017 and if that fails it picks the first unused input channel.
2019 Join 3 inputs (with properly set channel layouts):
2021 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2024 Build a 5.1 output from 6 single-channel streams:
2026 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2027 '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'
2033 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2035 To enable compilation of this filter you need to configure FFmpeg with
2036 @code{--enable-ladspa}.
2040 Specifies the name of LADSPA plugin library to load. If the environment
2041 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2042 each one of the directories specified by the colon separated list in
2043 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2044 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2045 @file{/usr/lib/ladspa/}.
2048 Specifies the plugin within the library. Some libraries contain only
2049 one plugin, but others contain many of them. If this is not set filter
2050 will list all available plugins within the specified library.
2053 Set the '|' separated list of controls which are zero or more floating point
2054 values that determine the behavior of the loaded plugin (for example delay,
2056 Controls need to be defined using the following syntax:
2057 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2058 @var{valuei} is the value set on the @var{i}-th control.
2059 Alternatively they can be also defined using the following syntax:
2060 @var{value0}|@var{value1}|@var{value2}|..., where
2061 @var{valuei} is the value set on the @var{i}-th control.
2062 If @option{controls} is set to @code{help}, all available controls and
2063 their valid ranges are printed.
2065 @item sample_rate, s
2066 Specify the sample rate, default to 44100. Only used if plugin have
2070 Set the number of samples per channel per each output frame, default
2071 is 1024. Only used if plugin have zero inputs.
2074 Set the minimum duration of the sourced audio. See
2075 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2076 for the accepted syntax.
2077 Note that the resulting duration may be greater than the specified duration,
2078 as the generated audio is always cut at the end of a complete frame.
2079 If not specified, or the expressed duration is negative, the audio is
2080 supposed to be generated forever.
2081 Only used if plugin have zero inputs.
2085 @subsection Examples
2089 List all available plugins within amp (LADSPA example plugin) library:
2095 List all available controls and their valid ranges for @code{vcf_notch}
2096 plugin from @code{VCF} library:
2098 ladspa=f=vcf:p=vcf_notch:c=help
2102 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2105 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2109 Add reverberation to the audio using TAP-plugins
2110 (Tom's Audio Processing plugins):
2112 ladspa=file=tap_reverb:tap_reverb
2116 Generate white noise, with 0.2 amplitude:
2118 ladspa=file=cmt:noise_source_white:c=c0=.2
2122 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2123 @code{C* Audio Plugin Suite} (CAPS) library:
2125 ladspa=file=caps:Click:c=c1=20'
2129 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2131 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2135 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2136 @code{SWH Plugins} collection:
2138 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2142 Attenuate low frequencies using Multiband EQ from Steve Harris
2143 @code{SWH Plugins} collection:
2145 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2149 @subsection Commands
2151 This filter supports the following commands:
2154 Modify the @var{N}-th control value.
2156 If the specified value is not valid, it is ignored and prior one is kept.
2161 Apply a low-pass filter with 3dB point frequency.
2162 The filter can be either single-pole or double-pole (the default).
2163 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2165 The filter accepts the following options:
2169 Set frequency in Hz. Default is 500.
2172 Set number of poles. Default is 2.
2175 Set method to specify band-width of filter.
2188 Specify the band-width of a filter in width_type units.
2189 Applies only to double-pole filter.
2190 The default is 0.707q and gives a Butterworth response.
2196 Mix channels with specific gain levels. The filter accepts the output
2197 channel layout followed by a set of channels definitions.
2199 This filter is also designed to efficiently remap the channels of an audio
2202 The filter accepts parameters of the form:
2203 "@var{l}|@var{outdef}|@var{outdef}|..."
2207 output channel layout or number of channels
2210 output channel specification, of the form:
2211 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
2214 output channel to define, either a channel name (FL, FR, etc.) or a channel
2215 number (c0, c1, etc.)
2218 multiplicative coefficient for the channel, 1 leaving the volume unchanged
2221 input channel to use, see out_name for details; it is not possible to mix
2222 named and numbered input channels
2225 If the `=' in a channel specification is replaced by `<', then the gains for
2226 that specification will be renormalized so that the total is 1, thus
2227 avoiding clipping noise.
2229 @subsection Mixing examples
2231 For example, if you want to down-mix from stereo to mono, but with a bigger
2232 factor for the left channel:
2234 pan=1c|c0=0.9*c0+0.1*c1
2237 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
2238 7-channels surround:
2240 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
2243 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
2244 that should be preferred (see "-ac" option) unless you have very specific
2247 @subsection Remapping examples
2249 The channel remapping will be effective if, and only if:
2252 @item gain coefficients are zeroes or ones,
2253 @item only one input per channel output,
2256 If all these conditions are satisfied, the filter will notify the user ("Pure
2257 channel mapping detected"), and use an optimized and lossless method to do the
2260 For example, if you have a 5.1 source and want a stereo audio stream by
2261 dropping the extra channels:
2263 pan="stereo| c0=FL | c1=FR"
2266 Given the same source, you can also switch front left and front right channels
2267 and keep the input channel layout:
2269 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
2272 If the input is a stereo audio stream, you can mute the front left channel (and
2273 still keep the stereo channel layout) with:
2278 Still with a stereo audio stream input, you can copy the right channel in both
2279 front left and right:
2281 pan="stereo| c0=FR | c1=FR"
2286 ReplayGain scanner filter. This filter takes an audio stream as an input and
2287 outputs it unchanged.
2288 At end of filtering it displays @code{track_gain} and @code{track_peak}.
2292 Convert the audio sample format, sample rate and channel layout. It is
2293 not meant to be used directly.
2296 Apply time-stretching and pitch-shifting with librubberband.
2298 The filter accepts the following options:
2302 Set tempo scale factor.
2305 Set pitch scale factor.
2308 Set transients detector.
2309 Possible values are:
2318 Possible values are:
2327 Possible values are:
2334 Set processing window size.
2335 Possible values are:
2344 Possible values are:
2351 Enable formant preservation when shift pitching.
2352 Possible values are:
2360 Possible values are:
2369 Possible values are:
2376 @section sidechaincompress
2378 This filter acts like normal compressor but has the ability to compress
2379 detected signal using second input signal.
2380 It needs two input streams and returns one output stream.
2381 First input stream will be processed depending on second stream signal.
2382 The filtered signal then can be filtered with other filters in later stages of
2383 processing. See @ref{pan} and @ref{amerge} filter.
2385 The filter accepts the following options:
2389 If a signal of second stream raises above this level it will affect the gain
2390 reduction of first stream.
2391 By default is 0.125. Range is between 0.00097563 and 1.
2394 Set a ratio about which the signal is reduced. 1:2 means that if the level
2395 raised 4dB above the threshold, it will be only 2dB above after the reduction.
2396 Default is 2. Range is between 1 and 20.
2399 Amount of milliseconds the signal has to rise above the threshold before gain
2400 reduction starts. Default is 20. Range is between 0.01 and 2000.
2403 Amount of milliseconds the signal has to fall below the threshold before
2404 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
2407 Set the amount by how much signal will be amplified after processing.
2408 Default is 2. Range is from 1 and 64.
2411 Curve the sharp knee around the threshold to enter gain reduction more softly.
2412 Default is 2.82843. Range is between 1 and 8.
2415 Choose if the @code{average} level between all channels of side-chain stream
2416 or the louder(@code{maximum}) channel of side-chain stream affects the
2417 reduction. Default is @code{average}.
2420 Should the exact signal be taken in case of @code{peak} or an RMS one in case
2421 of @code{rms}. Default is @code{rms} which is mainly smoother.
2424 @subsection Examples
2428 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
2429 depending on the signal of 2nd input and later compressed signal to be
2430 merged with 2nd input:
2432 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
2436 @section silencedetect
2438 Detect silence in an audio stream.
2440 This filter logs a message when it detects that the input audio volume is less
2441 or equal to a noise tolerance value for a duration greater or equal to the
2442 minimum detected noise duration.
2444 The printed times and duration are expressed in seconds.
2446 The filter accepts the following options:
2450 Set silence duration until notification (default is 2 seconds).
2453 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
2454 specified value) or amplitude ratio. Default is -60dB, or 0.001.
2457 @subsection Examples
2461 Detect 5 seconds of silence with -50dB noise tolerance:
2463 silencedetect=n=-50dB:d=5
2467 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
2468 tolerance in @file{silence.mp3}:
2470 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
2474 @section silenceremove
2476 Remove silence from the beginning, middle or end of the audio.
2478 The filter accepts the following options:
2482 This value is used to indicate if audio should be trimmed at beginning of
2483 the audio. A value of zero indicates no silence should be trimmed from the
2484 beginning. When specifying a non-zero value, it trims audio up until it
2485 finds non-silence. Normally, when trimming silence from beginning of audio
2486 the @var{start_periods} will be @code{1} but it can be increased to higher
2487 values to trim all audio up to specific count of non-silence periods.
2488 Default value is @code{0}.
2490 @item start_duration
2491 Specify the amount of time that non-silence must be detected before it stops
2492 trimming audio. By increasing the duration, bursts of noises can be treated
2493 as silence and trimmed off. Default value is @code{0}.
2495 @item start_threshold
2496 This indicates what sample value should be treated as silence. For digital
2497 audio, a value of @code{0} may be fine but for audio recorded from analog,
2498 you may wish to increase the value to account for background noise.
2499 Can be specified in dB (in case "dB" is appended to the specified value)
2500 or amplitude ratio. Default value is @code{0}.
2503 Set the count for trimming silence from the end of audio.
2504 To remove silence from the middle of a file, specify a @var{stop_periods}
2505 that is negative. This value is then treated as a positive value and is
2506 used to indicate the effect should restart processing as specified by
2507 @var{start_periods}, making it suitable for removing periods of silence
2508 in the middle of the audio.
2509 Default value is @code{0}.
2512 Specify a duration of silence that must exist before audio is not copied any
2513 more. By specifying a higher duration, silence that is wanted can be left in
2515 Default value is @code{0}.
2517 @item stop_threshold
2518 This is the same as @option{start_threshold} but for trimming silence from
2520 Can be specified in dB (in case "dB" is appended to the specified value)
2521 or amplitude ratio. Default value is @code{0}.
2524 This indicate that @var{stop_duration} length of audio should be left intact
2525 at the beginning of each period of silence.
2526 For example, if you want to remove long pauses between words but do not want
2527 to remove the pauses completely. Default value is @code{0}.
2531 @subsection Examples
2535 The following example shows how this filter can be used to start a recording
2536 that does not contain the delay at the start which usually occurs between
2537 pressing the record button and the start of the performance:
2539 silenceremove=1:5:0.02
2543 @section stereotools
2545 This filter has some handy utilities to manage stereo signals, for converting
2546 M/S stereo recordings to L/R signal while having control over the parameters
2547 or spreading the stereo image of master track.
2549 The filter accepts the following options:
2553 Set input level before filtering for both channels. Defaults is 1.
2554 Allowed range is from 0.015625 to 64.
2557 Set output level after filtering for both channels. Defaults is 1.
2558 Allowed range is from 0.015625 to 64.
2561 Set input balance between both channels. Default is 0.
2562 Allowed range is from -1 to 1.
2565 Set output balance between both channels. Default is 0.
2566 Allowed range is from -1 to 1.
2569 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
2570 clipping. Disabled by default.
2573 Mute the left channel. Disabled by default.
2576 Mute the right channel. Disabled by default.
2579 Change the phase of the left channel. Disabled by default.
2582 Change the phase of the right channel. Disabled by default.
2585 Set stereo mode. Available values are:
2589 Left/Right to Left/Right, this is default.
2592 Left/Right to Mid/Side.
2595 Mid/Side to Left/Right.
2598 Left/Right to Left/Left.
2601 Left/Right to Right/Right.
2604 Left/Right to Left + Right.
2607 Left/Right to Right/Left.
2611 Set level of side signal. Default is 1.
2612 Allowed range is from 0.015625 to 64.
2615 Set balance of side signal. Default is 0.
2616 Allowed range is from -1 to 1.
2619 Set level of the middle signal. Default is 1.
2620 Allowed range is from 0.015625 to 64.
2623 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
2626 Set stereo base between mono and inversed channels. Default is 0.
2627 Allowed range is from -1 to 1.
2630 Set delay in milliseconds how much to delay left from right channel and
2631 vice versa. Default is 0. Allowed range is from -20 to 20.
2634 Set S/C level. Default is 1. Allowed range is from 1 to 100.
2637 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
2640 @section stereowiden
2642 This filter enhance the stereo effect by suppressing signal common to both
2643 channels and by delaying the signal of left into right and vice versa,
2644 thereby widening the stereo effect.
2646 The filter accepts the following options:
2650 Time in milliseconds of the delay of left signal into right and vice versa.
2651 Default is 20 milliseconds.
2654 Amount of gain in delayed signal into right and vice versa. Gives a delay
2655 effect of left signal in right output and vice versa which gives widening
2656 effect. Default is 0.3.
2659 Cross feed of left into right with inverted phase. This helps in suppressing
2660 the mono. If the value is 1 it will cancel all the signal common to both
2661 channels. Default is 0.3.
2664 Set level of input signal of original channel. Default is 0.8.
2669 Boost or cut treble (upper) frequencies of the audio using a two-pole
2670 shelving filter with a response similar to that of a standard
2671 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
2673 The filter accepts the following options:
2677 Give the gain at whichever is the lower of ~22 kHz and the
2678 Nyquist frequency. Its useful range is about -20 (for a large cut)
2679 to +20 (for a large boost). Beware of clipping when using a positive gain.
2682 Set the filter's central frequency and so can be used
2683 to extend or reduce the frequency range to be boosted or cut.
2684 The default value is @code{3000} Hz.
2687 Set method to specify band-width of filter.
2700 Determine how steep is the filter's shelf transition.
2705 Sinusoidal amplitude modulation.
2707 The filter accepts the following options:
2711 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
2712 (20 Hz or lower) will result in a tremolo effect.
2713 This filter may also be used as a ring modulator by specifying
2714 a modulation frequency higher than 20 Hz.
2715 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
2718 Depth of modulation as a percentage. Range is 0.0 - 1.0.
2719 Default value is 0.5.
2724 Adjust the input audio volume.
2726 It accepts the following parameters:
2730 Set audio volume expression.
2732 Output values are clipped to the maximum value.
2734 The output audio volume is given by the relation:
2736 @var{output_volume} = @var{volume} * @var{input_volume}
2739 The default value for @var{volume} is "1.0".
2742 This parameter represents the mathematical precision.
2744 It determines which input sample formats will be allowed, which affects the
2745 precision of the volume scaling.
2749 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
2751 32-bit floating-point; this limits input sample format to FLT. (default)
2753 64-bit floating-point; this limits input sample format to DBL.
2757 Choose the behaviour on encountering ReplayGain side data in input frames.
2761 Remove ReplayGain side data, ignoring its contents (the default).
2764 Ignore ReplayGain side data, but leave it in the frame.
2767 Prefer the track gain, if present.
2770 Prefer the album gain, if present.
2773 @item replaygain_preamp
2774 Pre-amplification gain in dB to apply to the selected replaygain gain.
2776 Default value for @var{replaygain_preamp} is 0.0.
2779 Set when the volume expression is evaluated.
2781 It accepts the following values:
2784 only evaluate expression once during the filter initialization, or
2785 when the @samp{volume} command is sent
2788 evaluate expression for each incoming frame
2791 Default value is @samp{once}.
2794 The volume expression can contain the following parameters.
2798 frame number (starting at zero)
2801 @item nb_consumed_samples
2802 number of samples consumed by the filter
2804 number of samples in the current frame
2806 original frame position in the file
2812 PTS at start of stream
2814 time at start of stream
2820 last set volume value
2823 Note that when @option{eval} is set to @samp{once} only the
2824 @var{sample_rate} and @var{tb} variables are available, all other
2825 variables will evaluate to NAN.
2827 @subsection Commands
2829 This filter supports the following commands:
2832 Modify the volume expression.
2833 The command accepts the same syntax of the corresponding option.
2835 If the specified expression is not valid, it is kept at its current
2837 @item replaygain_noclip
2838 Prevent clipping by limiting the gain applied.
2840 Default value for @var{replaygain_noclip} is 1.
2844 @subsection Examples
2848 Halve the input audio volume:
2852 volume=volume=-6.0206dB
2855 In all the above example the named key for @option{volume} can be
2856 omitted, for example like in:
2862 Increase input audio power by 6 decibels using fixed-point precision:
2864 volume=volume=6dB:precision=fixed
2868 Fade volume after time 10 with an annihilation period of 5 seconds:
2870 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
2874 @section volumedetect
2876 Detect the volume of the input video.
2878 The filter has no parameters. The input is not modified. Statistics about
2879 the volume will be printed in the log when the input stream end is reached.
2881 In particular it will show the mean volume (root mean square), maximum
2882 volume (on a per-sample basis), and the beginning of a histogram of the
2883 registered volume values (from the maximum value to a cumulated 1/1000 of
2886 All volumes are in decibels relative to the maximum PCM value.
2888 @subsection Examples
2890 Here is an excerpt of the output:
2892 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
2893 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
2894 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
2895 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
2896 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
2897 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
2898 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
2899 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
2900 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
2906 The mean square energy is approximately -27 dB, or 10^-2.7.
2908 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
2910 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
2913 In other words, raising the volume by +4 dB does not cause any clipping,
2914 raising it by +5 dB causes clipping for 6 samples, etc.
2916 @c man end AUDIO FILTERS
2918 @chapter Audio Sources
2919 @c man begin AUDIO SOURCES
2921 Below is a description of the currently available audio sources.
2925 Buffer audio frames, and make them available to the filter chain.
2927 This source is mainly intended for a programmatic use, in particular
2928 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
2930 It accepts the following parameters:
2934 The timebase which will be used for timestamps of submitted frames. It must be
2935 either a floating-point number or in @var{numerator}/@var{denominator} form.
2938 The sample rate of the incoming audio buffers.
2941 The sample format of the incoming audio buffers.
2942 Either a sample format name or its corresponding integer representation from
2943 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
2945 @item channel_layout
2946 The channel layout of the incoming audio buffers.
2947 Either a channel layout name from channel_layout_map in
2948 @file{libavutil/channel_layout.c} or its corresponding integer representation
2949 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
2952 The number of channels of the incoming audio buffers.
2953 If both @var{channels} and @var{channel_layout} are specified, then they
2958 @subsection Examples
2961 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
2964 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
2965 Since the sample format with name "s16p" corresponds to the number
2966 6 and the "stereo" channel layout corresponds to the value 0x3, this is
2969 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
2974 Generate an audio signal specified by an expression.
2976 This source accepts in input one or more expressions (one for each
2977 channel), which are evaluated and used to generate a corresponding
2980 This source accepts the following options:
2984 Set the '|'-separated expressions list for each separate channel. In case the
2985 @option{channel_layout} option is not specified, the selected channel layout
2986 depends on the number of provided expressions. Otherwise the last
2987 specified expression is applied to the remaining output channels.
2989 @item channel_layout, c
2990 Set the channel layout. The number of channels in the specified layout
2991 must be equal to the number of specified expressions.
2994 Set the minimum duration of the sourced audio. See
2995 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2996 for the accepted syntax.
2997 Note that the resulting duration may be greater than the specified
2998 duration, as the generated audio is always cut at the end of a
3001 If not specified, or the expressed duration is negative, the audio is
3002 supposed to be generated forever.
3005 Set the number of samples per channel per each output frame,
3008 @item sample_rate, s
3009 Specify the sample rate, default to 44100.
3012 Each expression in @var{exprs} can contain the following constants:
3016 number of the evaluated sample, starting from 0
3019 time of the evaluated sample expressed in seconds, starting from 0
3026 @subsection Examples
3036 Generate a sin signal with frequency of 440 Hz, set sample rate to
3039 aevalsrc="sin(440*2*PI*t):s=8000"
3043 Generate a two channels signal, specify the channel layout (Front
3044 Center + Back Center) explicitly:
3046 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
3050 Generate white noise:
3052 aevalsrc="-2+random(0)"
3056 Generate an amplitude modulated signal:
3058 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
3062 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
3064 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
3071 The null audio source, return unprocessed audio frames. It is mainly useful
3072 as a template and to be employed in analysis / debugging tools, or as
3073 the source for filters which ignore the input data (for example the sox
3076 This source accepts the following options:
3080 @item channel_layout, cl
3082 Specifies the channel layout, and can be either an integer or a string
3083 representing a channel layout. The default value of @var{channel_layout}
3086 Check the channel_layout_map definition in
3087 @file{libavutil/channel_layout.c} for the mapping between strings and
3088 channel layout values.
3090 @item sample_rate, r
3091 Specifies the sample rate, and defaults to 44100.
3094 Set the number of samples per requested frames.
3098 @subsection Examples
3102 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
3104 anullsrc=r=48000:cl=4
3108 Do the same operation with a more obvious syntax:
3110 anullsrc=r=48000:cl=mono
3114 All the parameters need to be explicitly defined.
3118 Synthesize a voice utterance using the libflite library.
3120 To enable compilation of this filter you need to configure FFmpeg with
3121 @code{--enable-libflite}.
3123 Note that the flite library is not thread-safe.
3125 The filter accepts the following options:
3130 If set to 1, list the names of the available voices and exit
3131 immediately. Default value is 0.
3134 Set the maximum number of samples per frame. Default value is 512.
3137 Set the filename containing the text to speak.
3140 Set the text to speak.
3143 Set the voice to use for the speech synthesis. Default value is
3144 @code{kal}. See also the @var{list_voices} option.
3147 @subsection Examples
3151 Read from file @file{speech.txt}, and synthesize the text using the
3152 standard flite voice:
3154 flite=textfile=speech.txt
3158 Read the specified text selecting the @code{slt} voice:
3160 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3164 Input text to ffmpeg:
3166 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3170 Make @file{ffplay} speak the specified text, using @code{flite} and
3171 the @code{lavfi} device:
3173 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
3177 For more information about libflite, check:
3178 @url{http://www.speech.cs.cmu.edu/flite/}
3182 Generate an audio signal made of a sine wave with amplitude 1/8.
3184 The audio signal is bit-exact.
3186 The filter accepts the following options:
3191 Set the carrier frequency. Default is 440 Hz.
3193 @item beep_factor, b
3194 Enable a periodic beep every second with frequency @var{beep_factor} times
3195 the carrier frequency. Default is 0, meaning the beep is disabled.
3197 @item sample_rate, r
3198 Specify the sample rate, default is 44100.
3201 Specify the duration of the generated audio stream.
3203 @item samples_per_frame
3204 Set the number of samples per output frame.
3206 The expression can contain the following constants:
3210 The (sequential) number of the output audio frame, starting from 0.
3213 The PTS (Presentation TimeStamp) of the output audio frame,
3214 expressed in @var{TB} units.
3217 The PTS of the output audio frame, expressed in seconds.
3220 The timebase of the output audio frames.
3223 Default is @code{1024}.
3226 @subsection Examples
3231 Generate a simple 440 Hz sine wave:
3237 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
3241 sine=frequency=220:beep_factor=4:duration=5
3245 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
3248 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
3252 @c man end AUDIO SOURCES
3254 @chapter Audio Sinks
3255 @c man begin AUDIO SINKS
3257 Below is a description of the currently available audio sinks.
3259 @section abuffersink
3261 Buffer audio frames, and make them available to the end of filter chain.
3263 This sink is mainly intended for programmatic use, in particular
3264 through the interface defined in @file{libavfilter/buffersink.h}
3265 or the options system.
3267 It accepts a pointer to an AVABufferSinkContext structure, which
3268 defines the incoming buffers' formats, to be passed as the opaque
3269 parameter to @code{avfilter_init_filter} for initialization.
3272 Null audio sink; do absolutely nothing with the input audio. It is
3273 mainly useful as a template and for use in analysis / debugging
3276 @c man end AUDIO SINKS
3278 @chapter Video Filters
3279 @c man begin VIDEO FILTERS
3281 When you configure your FFmpeg build, you can disable any of the
3282 existing filters using @code{--disable-filters}.
3283 The configure output will show the video filters included in your
3286 Below is a description of the currently available video filters.
3288 @section alphaextract
3290 Extract the alpha component from the input as a grayscale video. This
3291 is especially useful with the @var{alphamerge} filter.
3295 Add or replace the alpha component of the primary input with the
3296 grayscale value of a second input. This is intended for use with
3297 @var{alphaextract} to allow the transmission or storage of frame
3298 sequences that have alpha in a format that doesn't support an alpha
3301 For example, to reconstruct full frames from a normal YUV-encoded video
3302 and a separate video created with @var{alphaextract}, you might use:
3304 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
3307 Since this filter is designed for reconstruction, it operates on frame
3308 sequences without considering timestamps, and terminates when either
3309 input reaches end of stream. This will cause problems if your encoding
3310 pipeline drops frames. If you're trying to apply an image as an
3311 overlay to a video stream, consider the @var{overlay} filter instead.
3315 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
3316 and libavformat to work. On the other hand, it is limited to ASS (Advanced
3317 Substation Alpha) subtitles files.
3319 This filter accepts the following option in addition to the common options from
3320 the @ref{subtitles} filter:
3324 Set the shaping engine
3326 Available values are:
3329 The default libass shaping engine, which is the best available.
3331 Fast, font-agnostic shaper that can do only substitutions
3333 Slower shaper using OpenType for substitutions and positioning
3336 The default is @code{auto}.
3340 Apply an Adaptive Temporal Averaging Denoiser to the video input.
3342 The filter accepts the following options:
3346 Set threshold A for 1st plane. Default is 0.02.
3347 Valid range is 0 to 0.3.
3350 Set threshold B for 1st plane. Default is 0.04.
3351 Valid range is 0 to 5.
3354 Set threshold A for 2nd plane. Default is 0.02.
3355 Valid range is 0 to 0.3.
3358 Set threshold B for 2nd plane. Default is 0.04.
3359 Valid range is 0 to 5.
3362 Set threshold A for 3rd plane. Default is 0.02.
3363 Valid range is 0 to 0.3.
3366 Set threshold B for 3rd plane. Default is 0.04.
3367 Valid range is 0 to 5.
3369 Threshold A is designed to react on abrupt changes in the input signal and
3370 threshold B is designed to react on continuous changes in the input signal.
3373 Set number of frames filter will use for averaging. Default is 33. Must be odd
3374 number in range [5, 129].
3379 Compute the bounding box for the non-black pixels in the input frame
3382 This filter computes the bounding box containing all the pixels with a
3383 luminance value greater than the minimum allowed value.
3384 The parameters describing the bounding box are printed on the filter
3387 The filter accepts the following option:
3391 Set the minimal luminance value. Default is @code{16}.
3394 @section blackdetect
3396 Detect video intervals that are (almost) completely black. Can be
3397 useful to detect chapter transitions, commercials, or invalid
3398 recordings. Output lines contains the time for the start, end and
3399 duration of the detected black interval expressed in seconds.
3401 In order to display the output lines, you need to set the loglevel at
3402 least to the AV_LOG_INFO value.
3404 The filter accepts the following options:
3407 @item black_min_duration, d
3408 Set the minimum detected black duration expressed in seconds. It must
3409 be a non-negative floating point number.
3411 Default value is 2.0.
3413 @item picture_black_ratio_th, pic_th
3414 Set the threshold for considering a picture "black".
3415 Express the minimum value for the ratio:
3417 @var{nb_black_pixels} / @var{nb_pixels}
3420 for which a picture is considered black.
3421 Default value is 0.98.
3423 @item pixel_black_th, pix_th
3424 Set the threshold for considering a pixel "black".
3426 The threshold expresses the maximum pixel luminance value for which a
3427 pixel is considered "black". The provided value is scaled according to
3428 the following equation:
3430 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
3433 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
3434 the input video format, the range is [0-255] for YUV full-range
3435 formats and [16-235] for YUV non full-range formats.
3437 Default value is 0.10.
3440 The following example sets the maximum pixel threshold to the minimum
3441 value, and detects only black intervals of 2 or more seconds:
3443 blackdetect=d=2:pix_th=0.00
3448 Detect frames that are (almost) completely black. Can be useful to
3449 detect chapter transitions or commercials. Output lines consist of
3450 the frame number of the detected frame, the percentage of blackness,
3451 the position in the file if known or -1 and the timestamp in seconds.
3453 In order to display the output lines, you need to set the loglevel at
3454 least to the AV_LOG_INFO value.
3456 It accepts the following parameters:
3461 The percentage of the pixels that have to be below the threshold; it defaults to
3464 @item threshold, thresh
3465 The threshold below which a pixel value is considered black; it defaults to
3470 @section blend, tblend
3472 Blend two video frames into each other.
3474 The @code{blend} filter takes two input streams and outputs one
3475 stream, the first input is the "top" layer and second input is
3476 "bottom" layer. Output terminates when shortest input terminates.
3478 The @code{tblend} (time blend) filter takes two consecutive frames
3479 from one single stream, and outputs the result obtained by blending
3480 the new frame on top of the old frame.
3482 A description of the accepted options follows.
3490 Set blend mode for specific pixel component or all pixel components in case
3491 of @var{all_mode}. Default value is @code{normal}.
3493 Available values for component modes are:
3531 Set blend opacity for specific pixel component or all pixel components in case
3532 of @var{all_opacity}. Only used in combination with pixel component blend modes.
3539 Set blend expression for specific pixel component or all pixel components in case
3540 of @var{all_expr}. Note that related mode options will be ignored if those are set.
3542 The expressions can use the following variables:
3546 The sequential number of the filtered frame, starting from @code{0}.
3550 the coordinates of the current sample
3554 the width and height of currently filtered plane
3558 Width and height scale depending on the currently filtered plane. It is the
3559 ratio between the corresponding luma plane number of pixels and the current
3560 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
3561 @code{0.5,0.5} for chroma planes.
3564 Time of the current frame, expressed in seconds.
3567 Value of pixel component at current location for first video frame (top layer).
3570 Value of pixel component at current location for second video frame (bottom layer).
3574 Force termination when the shortest input terminates. Default is
3575 @code{0}. This option is only defined for the @code{blend} filter.
3578 Continue applying the last bottom frame after the end of the stream. A value of
3579 @code{0} disable the filter after the last frame of the bottom layer is reached.
3580 Default is @code{1}. This option is only defined for the @code{blend} filter.
3583 @subsection Examples
3587 Apply transition from bottom layer to top layer in first 10 seconds:
3589 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
3593 Apply 1x1 checkerboard effect:
3595 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
3599 Apply uncover left effect:
3601 blend=all_expr='if(gte(N*SW+X,W),A,B)'
3605 Apply uncover down effect:
3607 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
3611 Apply uncover up-left effect:
3613 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
3617 Display differences between the current and the previous frame:
3619 tblend=all_mode=difference128
3625 Apply a boxblur algorithm to the input video.
3627 It accepts the following parameters:
3631 @item luma_radius, lr
3632 @item luma_power, lp
3633 @item chroma_radius, cr
3634 @item chroma_power, cp
3635 @item alpha_radius, ar
3636 @item alpha_power, ap
3640 A description of the accepted options follows.
3643 @item luma_radius, lr
3644 @item chroma_radius, cr
3645 @item alpha_radius, ar
3646 Set an expression for the box radius in pixels used for blurring the
3647 corresponding input plane.
3649 The radius value must be a non-negative number, and must not be
3650 greater than the value of the expression @code{min(w,h)/2} for the
3651 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
3654 Default value for @option{luma_radius} is "2". If not specified,
3655 @option{chroma_radius} and @option{alpha_radius} default to the
3656 corresponding value set for @option{luma_radius}.
3658 The expressions can contain the following constants:
3662 The input width and height in pixels.
3666 The input chroma image width and height in pixels.
3670 The horizontal and vertical chroma subsample values. For example, for the
3671 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
3674 @item luma_power, lp
3675 @item chroma_power, cp
3676 @item alpha_power, ap
3677 Specify how many times the boxblur filter is applied to the
3678 corresponding plane.
3680 Default value for @option{luma_power} is 2. If not specified,
3681 @option{chroma_power} and @option{alpha_power} default to the
3682 corresponding value set for @option{luma_power}.
3684 A value of 0 will disable the effect.
3687 @subsection Examples
3691 Apply a boxblur filter with the luma, chroma, and alpha radii
3694 boxblur=luma_radius=2:luma_power=1
3699 Set the luma radius to 2, and alpha and chroma radius to 0:
3701 boxblur=2:1:cr=0:ar=0
3705 Set the luma and chroma radii to a fraction of the video dimension:
3707 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
3712 YUV colorspace color/chroma keying.
3714 The filter accepts the following options:
3718 The color which will be replaced with transparency.
3721 Similarity percentage with the key color.
3723 0.01 matches only the exact key color, while 1.0 matches everything.
3728 0.0 makes pixels either fully transparent, or not transparent at all.
3730 Higher values result in semi-transparent pixels, with a higher transparency
3731 the more similar the pixels color is to the key color.
3734 Signals that the color passed is already in YUV instead of RGB.
3736 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
3737 This can be used to pass exact YUV values as hexadecimal numbers.
3740 @subsection Examples
3744 Make every green pixel in the input image transparent:
3746 ffmpeg -i input.png -vf chromakey=green out.png
3750 Overlay a greenscreen-video on top of a static black background.
3752 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
3758 Visualize information exported by some codecs.
3760 Some codecs can export information through frames using side-data or other
3761 means. For example, some MPEG based codecs export motion vectors through the
3762 @var{export_mvs} flag in the codec @option{flags2} option.
3764 The filter accepts the following option:
3768 Set motion vectors to visualize.
3770 Available flags for @var{mv} are:
3774 forward predicted MVs of P-frames
3776 forward predicted MVs of B-frames
3778 backward predicted MVs of B-frames
3782 @subsection Examples
3786 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
3788 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
3792 @section colorbalance
3793 Modify intensity of primary colors (red, green and blue) of input frames.
3795 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
3796 regions for the red-cyan, green-magenta or blue-yellow balance.
3798 A positive adjustment value shifts the balance towards the primary color, a negative
3799 value towards the complementary color.
3801 The filter accepts the following options:
3807 Adjust red, green and blue shadows (darkest pixels).
3812 Adjust red, green and blue midtones (medium pixels).
3817 Adjust red, green and blue highlights (brightest pixels).
3819 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3822 @subsection Examples
3826 Add red color cast to shadows:
3833 RGB colorspace color keying.
3835 The filter accepts the following options:
3839 The color which will be replaced with transparency.
3842 Similarity percentage with the key color.
3844 0.01 matches only the exact key color, while 1.0 matches everything.
3849 0.0 makes pixels either fully transparent, or not transparent at all.
3851 Higher values result in semi-transparent pixels, with a higher transparency
3852 the more similar the pixels color is to the key color.
3855 @subsection Examples
3859 Make every green pixel in the input image transparent:
3861 ffmpeg -i input.png -vf colorkey=green out.png
3865 Overlay a greenscreen-video on top of a static background image.
3867 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
3871 @section colorlevels
3873 Adjust video input frames using levels.
3875 The filter accepts the following options:
3882 Adjust red, green, blue and alpha input black point.
3883 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3889 Adjust red, green, blue and alpha input white point.
3890 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
3892 Input levels are used to lighten highlights (bright tones), darken shadows
3893 (dark tones), change the balance of bright and dark tones.
3899 Adjust red, green, blue and alpha output black point.
3900 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
3906 Adjust red, green, blue and alpha output white point.
3907 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
3909 Output levels allows manual selection of a constrained output level range.
3912 @subsection Examples
3916 Make video output darker:
3918 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
3924 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
3928 Make video output lighter:
3930 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
3934 Increase brightness:
3936 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
3940 @section colorchannelmixer
3942 Adjust video input frames by re-mixing color channels.
3944 This filter modifies a color channel by adding the values associated to
3945 the other channels of the same pixels. For example if the value to
3946 modify is red, the output value will be:
3948 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
3951 The filter accepts the following options:
3958 Adjust contribution of input red, green, blue and alpha channels for output red channel.
3959 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
3965 Adjust contribution of input red, green, blue and alpha channels for output green channel.
3966 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
3972 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
3973 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
3979 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
3980 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
3982 Allowed ranges for options are @code{[-2.0, 2.0]}.
3985 @subsection Examples
3989 Convert source to grayscale:
3991 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
3994 Simulate sepia tones:
3996 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
4000 @section colormatrix
4002 Convert color matrix.
4004 The filter accepts the following options:
4009 Specify the source and destination color matrix. Both values must be
4012 The accepted values are:
4028 For example to convert from BT.601 to SMPTE-240M, use the command:
4030 colormatrix=bt601:smpte240m
4035 Copy the input source unchanged to the output. This is mainly useful for
4040 Crop the input video to given dimensions.
4042 It accepts the following parameters:
4046 The width of the output video. It defaults to @code{iw}.
4047 This expression is evaluated only once during the filter
4048 configuration, or when the @samp{w} or @samp{out_w} command is sent.
4051 The height of the output video. It defaults to @code{ih}.
4052 This expression is evaluated only once during the filter
4053 configuration, or when the @samp{h} or @samp{out_h} command is sent.
4056 The horizontal position, in the input video, of the left edge of the output
4057 video. It defaults to @code{(in_w-out_w)/2}.
4058 This expression is evaluated per-frame.
4061 The vertical position, in the input video, of the top edge of the output video.
4062 It defaults to @code{(in_h-out_h)/2}.
4063 This expression is evaluated per-frame.
4066 If set to 1 will force the output display aspect ratio
4067 to be the same of the input, by changing the output sample aspect
4068 ratio. It defaults to 0.
4071 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
4072 expressions containing the following constants:
4077 The computed values for @var{x} and @var{y}. They are evaluated for
4082 The input width and height.
4086 These are the same as @var{in_w} and @var{in_h}.
4090 The output (cropped) width and height.
4094 These are the same as @var{out_w} and @var{out_h}.
4097 same as @var{iw} / @var{ih}
4100 input sample aspect ratio
4103 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
4107 horizontal and vertical chroma subsample values. For example for the
4108 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4111 The number of the input frame, starting from 0.
4114 the position in the file of the input frame, NAN if unknown
4117 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
4121 The expression for @var{out_w} may depend on the value of @var{out_h},
4122 and the expression for @var{out_h} may depend on @var{out_w}, but they
4123 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
4124 evaluated after @var{out_w} and @var{out_h}.
4126 The @var{x} and @var{y} parameters specify the expressions for the
4127 position of the top-left corner of the output (non-cropped) area. They
4128 are evaluated for each frame. If the evaluated value is not valid, it
4129 is approximated to the nearest valid value.
4131 The expression for @var{x} may depend on @var{y}, and the expression
4132 for @var{y} may depend on @var{x}.
4134 @subsection Examples
4138 Crop area with size 100x100 at position (12,34).
4143 Using named options, the example above becomes:
4145 crop=w=100:h=100:x=12:y=34
4149 Crop the central input area with size 100x100:
4155 Crop the central input area with size 2/3 of the input video:
4157 crop=2/3*in_w:2/3*in_h
4161 Crop the input video central square:
4168 Delimit the rectangle with the top-left corner placed at position
4169 100:100 and the right-bottom corner corresponding to the right-bottom
4170 corner of the input image.
4172 crop=in_w-100:in_h-100:100:100
4176 Crop 10 pixels from the left and right borders, and 20 pixels from
4177 the top and bottom borders
4179 crop=in_w-2*10:in_h-2*20
4183 Keep only the bottom right quarter of the input image:
4185 crop=in_w/2:in_h/2:in_w/2:in_h/2
4189 Crop height for getting Greek harmony:
4191 crop=in_w:1/PHI*in_w
4195 Apply trembling effect:
4197 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)
4201 Apply erratic camera effect depending on timestamp:
4203 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)"
4207 Set x depending on the value of y:
4209 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
4213 @subsection Commands
4215 This filter supports the following commands:
4221 Set width/height of the output video and the horizontal/vertical position
4223 The command accepts the same syntax of the corresponding option.
4225 If the specified expression is not valid, it is kept at its current
4231 Auto-detect the crop size.
4233 It calculates the necessary cropping parameters and prints the
4234 recommended parameters via the logging system. The detected dimensions
4235 correspond to the non-black area of the input video.
4237 It accepts the following parameters:
4242 Set higher black value threshold, which can be optionally specified
4243 from nothing (0) to everything (255 for 8bit based formats). An intensity
4244 value greater to the set value is considered non-black. It defaults to 24.
4245 You can also specify a value between 0.0 and 1.0 which will be scaled depending
4246 on the bitdepth of the pixel format.
4249 The value which the width/height should be divisible by. It defaults to
4250 16. The offset is automatically adjusted to center the video. Use 2 to
4251 get only even dimensions (needed for 4:2:2 video). 16 is best when
4252 encoding to most video codecs.
4254 @item reset_count, reset
4255 Set the counter that determines after how many frames cropdetect will
4256 reset the previously detected largest video area and start over to
4257 detect the current optimal crop area. Default value is 0.
4259 This can be useful when channel logos distort the video area. 0
4260 indicates 'never reset', and returns the largest area encountered during
4267 Apply color adjustments using curves.
4269 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
4270 component (red, green and blue) has its values defined by @var{N} key points
4271 tied from each other using a smooth curve. The x-axis represents the pixel
4272 values from the input frame, and the y-axis the new pixel values to be set for
4275 By default, a component curve is defined by the two points @var{(0;0)} and
4276 @var{(1;1)}. This creates a straight line where each original pixel value is
4277 "adjusted" to its own value, which means no change to the image.
4279 The filter allows you to redefine these two points and add some more. A new
4280 curve (using a natural cubic spline interpolation) will be define to pass
4281 smoothly through all these new coordinates. The new defined points needs to be
4282 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
4283 be in the @var{[0;1]} interval. If the computed curves happened to go outside
4284 the vector spaces, the values will be clipped accordingly.
4286 If there is no key point defined in @code{x=0}, the filter will automatically
4287 insert a @var{(0;0)} point. In the same way, if there is no key point defined
4288 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
4290 The filter accepts the following options:
4294 Select one of the available color presets. This option can be used in addition
4295 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
4296 options takes priority on the preset values.
4297 Available presets are:
4300 @item color_negative
4303 @item increase_contrast
4305 @item linear_contrast
4306 @item medium_contrast
4308 @item strong_contrast
4311 Default is @code{none}.
4313 Set the master key points. These points will define a second pass mapping. It
4314 is sometimes called a "luminance" or "value" mapping. It can be used with
4315 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
4316 post-processing LUT.
4318 Set the key points for the red component.
4320 Set the key points for the green component.
4322 Set the key points for the blue component.
4324 Set the key points for all components (not including master).
4325 Can be used in addition to the other key points component
4326 options. In this case, the unset component(s) will fallback on this
4327 @option{all} setting.
4329 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
4332 To avoid some filtergraph syntax conflicts, each key points list need to be
4333 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
4335 @subsection Examples
4339 Increase slightly the middle level of blue:
4341 curves=blue='0.5/0.58'
4347 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
4349 Here we obtain the following coordinates for each components:
4352 @code{(0;0.11) (0.42;0.51) (1;0.95)}
4354 @code{(0;0) (0.50;0.48) (1;1)}
4356 @code{(0;0.22) (0.49;0.44) (1;0.80)}
4360 The previous example can also be achieved with the associated built-in preset:
4362 curves=preset=vintage
4372 Use a Photoshop preset and redefine the points of the green component:
4374 curves=psfile='MyCurvesPresets/purple.acv':green='0.45/0.53'
4380 Denoise frames using 2D DCT (frequency domain filtering).
4382 This filter is not designed for real time.
4384 The filter accepts the following options:
4388 Set the noise sigma constant.
4390 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
4391 coefficient (absolute value) below this threshold with be dropped.
4393 If you need a more advanced filtering, see @option{expr}.
4395 Default is @code{0}.
4398 Set number overlapping pixels for each block. Since the filter can be slow, you
4399 may want to reduce this value, at the cost of a less effective filter and the
4400 risk of various artefacts.
4402 If the overlapping value doesn't permit processing the whole input width or
4403 height, a warning will be displayed and according borders won't be denoised.
4405 Default value is @var{blocksize}-1, which is the best possible setting.
4408 Set the coefficient factor expression.
4410 For each coefficient of a DCT block, this expression will be evaluated as a
4411 multiplier value for the coefficient.
4413 If this is option is set, the @option{sigma} option will be ignored.
4415 The absolute value of the coefficient can be accessed through the @var{c}
4419 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
4420 @var{blocksize}, which is the width and height of the processed blocks.
4422 The default value is @var{3} (8x8) and can be raised to @var{4} for a
4423 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
4424 on the speed processing. Also, a larger block size does not necessarily means a
4428 @subsection Examples
4430 Apply a denoise with a @option{sigma} of @code{4.5}:
4435 The same operation can be achieved using the expression system:
4437 dctdnoiz=e='gte(c, 4.5*3)'
4440 Violent denoise using a block size of @code{16x16}:
4447 Remove banding artifacts from input video.
4448 It works by replacing banded pixels with average value of referenced pixels.
4450 The filter accepts the following options:
4457 Set banding detection threshold for each plane. Default is 0.02.
4458 Valid range is 0.00003 to 0.5.
4459 If difference between current pixel and reference pixel is less than threshold,
4460 it will be considered as banded.
4463 Banding detection range in pixels. Default is 16. If positive, random number
4464 in range 0 to set value will be used. If negative, exact absolute value
4466 The range defines square of four pixels around current pixel.
4469 Set direction in radians from which four pixel will be compared. If positive,
4470 random direction from 0 to set direction will be picked. If negative, exact of
4471 absolute value will be picked. For example direction 0, -PI or -2*PI radians
4472 will pick only pixels on same row and -PI/2 will pick only pixels on same
4476 If enabled, current pixel is compared with average value of all four
4477 surrounding pixels. The default is enabled. If disabled current pixel is
4478 compared with all four surrounding pixels. The pixel is considered banded
4479 if only all four differences with surrounding pixels are less than threshold.
4485 Drop duplicated frames at regular intervals.
4487 The filter accepts the following options:
4491 Set the number of frames from which one will be dropped. Setting this to
4492 @var{N} means one frame in every batch of @var{N} frames will be dropped.
4493 Default is @code{5}.
4496 Set the threshold for duplicate detection. If the difference metric for a frame
4497 is less than or equal to this value, then it is declared as duplicate. Default
4501 Set scene change threshold. Default is @code{15}.
4505 Set the size of the x and y-axis blocks used during metric calculations.
4506 Larger blocks give better noise suppression, but also give worse detection of
4507 small movements. Must be a power of two. Default is @code{32}.
4510 Mark main input as a pre-processed input and activate clean source input
4511 stream. This allows the input to be pre-processed with various filters to help
4512 the metrics calculation while keeping the frame selection lossless. When set to
4513 @code{1}, the first stream is for the pre-processed input, and the second
4514 stream is the clean source from where the kept frames are chosen. Default is
4518 Set whether or not chroma is considered in the metric calculations. Default is
4524 Apply deflate effect to the video.
4526 This filter replaces the pixel by the local(3x3) average by taking into account
4527 only values lower than the pixel.
4529 It accepts the following options:
4536 Limit the maximum change for each plane, default is 65535.
4537 If 0, plane will remain unchanged.
4542 Remove judder produced by partially interlaced telecined content.
4544 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
4545 source was partially telecined content then the output of @code{pullup,dejudder}
4546 will have a variable frame rate. May change the recorded frame rate of the
4547 container. Aside from that change, this filter will not affect constant frame
4550 The option available in this filter is:
4554 Specify the length of the window over which the judder repeats.
4556 Accepts any integer greater than 1. Useful values are:
4560 If the original was telecined from 24 to 30 fps (Film to NTSC).
4563 If the original was telecined from 25 to 30 fps (PAL to NTSC).
4566 If a mixture of the two.
4569 The default is @samp{4}.
4574 Suppress a TV station logo by a simple interpolation of the surrounding
4575 pixels. Just set a rectangle covering the logo and watch it disappear
4576 (and sometimes something even uglier appear - your mileage may vary).
4578 It accepts the following parameters:
4583 Specify the top left corner coordinates of the logo. They must be
4588 Specify the width and height of the logo to clear. They must be
4592 Specify the thickness of the fuzzy edge of the rectangle (added to
4593 @var{w} and @var{h}). The default value is 1. This option is
4594 deprecated, setting higher values should no longer be necessary and
4598 When set to 1, a green rectangle is drawn on the screen to simplify
4599 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
4600 The default value is 0.
4602 The rectangle is drawn on the outermost pixels which will be (partly)
4603 replaced with interpolated values. The values of the next pixels
4604 immediately outside this rectangle in each direction will be used to
4605 compute the interpolated pixel values inside the rectangle.
4609 @subsection Examples
4613 Set a rectangle covering the area with top left corner coordinates 0,0
4614 and size 100x77, and a band of size 10:
4616 delogo=x=0:y=0:w=100:h=77:band=10
4623 Attempt to fix small changes in horizontal and/or vertical shift. This
4624 filter helps remove camera shake from hand-holding a camera, bumping a
4625 tripod, moving on a vehicle, etc.
4627 The filter accepts the following options:
4635 Specify a rectangular area where to limit the search for motion
4637 If desired the search for motion vectors can be limited to a
4638 rectangular area of the frame defined by its top left corner, width
4639 and height. These parameters have the same meaning as the drawbox
4640 filter which can be used to visualise the position of the bounding
4643 This is useful when simultaneous movement of subjects within the frame
4644 might be confused for camera motion by the motion vector search.
4646 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
4647 then the full frame is used. This allows later options to be set
4648 without specifying the bounding box for the motion vector search.
4650 Default - search the whole frame.
4654 Specify the maximum extent of movement in x and y directions in the
4655 range 0-64 pixels. Default 16.
4658 Specify how to generate pixels to fill blanks at the edge of the
4659 frame. Available values are:
4662 Fill zeroes at blank locations
4664 Original image at blank locations
4666 Extruded edge value at blank locations
4668 Mirrored edge at blank locations
4670 Default value is @samp{mirror}.
4673 Specify the blocksize to use for motion search. Range 4-128 pixels,
4677 Specify the contrast threshold for blocks. Only blocks with more than
4678 the specified contrast (difference between darkest and lightest
4679 pixels) will be considered. Range 1-255, default 125.
4682 Specify the search strategy. Available values are:
4685 Set exhaustive search
4687 Set less exhaustive search.
4689 Default value is @samp{exhaustive}.
4692 If set then a detailed log of the motion search is written to the
4696 If set to 1, specify using OpenCL capabilities, only available if
4697 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
4703 Apply an exact inverse of the telecine operation. It requires a predefined
4704 pattern specified using the pattern option which must be the same as that passed
4705 to the telecine filter.
4707 This filter accepts the following options:
4716 The default value is @code{top}.
4720 A string of numbers representing the pulldown pattern you wish to apply.
4721 The default value is @code{23}.
4724 A number representing position of the first frame with respect to the telecine
4725 pattern. This is to be used if the stream is cut. The default value is @code{0}.
4730 Apply dilation effect to the video.
4732 This filter replaces the pixel by the local(3x3) maximum.
4734 It accepts the following options:
4741 Limit the maximum change for each plane, default is 65535.
4742 If 0, plane will remain unchanged.
4745 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
4748 Flags to local 3x3 coordinates maps like this:
4757 Displace pixels as indicated by second and third input stream.
4759 It takes three input streams and outputs one stream, the first input is the
4760 source, and second and third input are displacement maps.
4762 The second input specifies how much to displace pixels along the
4763 x-axis, while the third input specifies how much to displace pixels
4765 If one of displacement map streams terminates, last frame from that
4766 displacement map will be used.
4768 Note that once generated, displacements maps can be reused over and over again.
4770 A description of the accepted options follows.
4774 Set displace behavior for pixels that are out of range.
4776 Available values are:
4779 Missing pixels are replaced by black pixels.
4782 Adjacent pixels will spread out to replace missing pixels.
4785 Out of range pixels are wrapped so they point to pixels of other side.
4787 Default is @samp{smear}.
4791 @subsection Examples
4795 Add ripple effect to rgb input of video size hd720:
4797 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
4801 Add wave effect to rgb input of video size hd720:
4803 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
4809 Draw a colored box on the input image.
4811 It accepts the following parameters:
4816 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
4820 The expressions which specify the width and height of the box; if 0 they are interpreted as
4821 the input width and height. It defaults to 0.
4824 Specify the color of the box to write. For the general syntax of this option,
4825 check the "Color" section in the ffmpeg-utils manual. If the special
4826 value @code{invert} is used, the box edge color is the same as the
4827 video with inverted luma.
4830 The expression which sets the thickness of the box edge. Default value is @code{3}.
4832 See below for the list of accepted constants.
4835 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
4836 following constants:
4840 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
4844 horizontal and vertical chroma subsample values. For example for the
4845 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4849 The input width and height.
4852 The input sample aspect ratio.
4856 The x and y offset coordinates where the box is drawn.
4860 The width and height of the drawn box.
4863 The thickness of the drawn box.
4865 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
4866 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
4870 @subsection Examples
4874 Draw a black box around the edge of the input image:
4880 Draw a box with color red and an opacity of 50%:
4882 drawbox=10:20:200:60:red@@0.5
4885 The previous example can be specified as:
4887 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
4891 Fill the box with pink color:
4893 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
4897 Draw a 2-pixel red 2.40:1 mask:
4899 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
4903 @section drawgraph, adrawgraph
4905 Draw a graph using input video or audio metadata.
4907 It accepts the following parameters:
4911 Set 1st frame metadata key from which metadata values will be used to draw a graph.
4914 Set 1st foreground color expression.
4917 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
4920 Set 2nd foreground color expression.
4923 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
4926 Set 3rd foreground color expression.
4929 Set 4th frame metadata key from which metadata values will be used to draw a graph.
4932 Set 4th foreground color expression.
4935 Set minimal value of metadata value.
4938 Set maximal value of metadata value.
4941 Set graph background color. Default is white.
4946 Available values for mode is:
4953 Default is @code{line}.
4958 Available values for slide is:
4961 Draw new frame when right border is reached.
4964 Replace old columns with new ones.
4967 Scroll from right to left.
4970 Scroll from left to right.
4973 Default is @code{frame}.
4976 Set size of graph video. For the syntax of this option, check the
4977 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
4978 The default value is @code{900x256}.
4980 The foreground color expressions can use the following variables:
4983 Minimal value of metadata value.
4986 Maximal value of metadata value.
4989 Current metadata key value.
4992 The color is defined as 0xAABBGGRR.
4995 Example using metadata from @ref{signalstats} filter:
4997 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
5000 Example using metadata from @ref{ebur128} filter:
5002 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
5007 Draw a grid on the input image.
5009 It accepts the following parameters:
5014 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
5018 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
5019 input width and height, respectively, minus @code{thickness}, so image gets
5020 framed. Default to 0.
5023 Specify the color of the grid. For the general syntax of this option,
5024 check the "Color" section in the ffmpeg-utils manual. If the special
5025 value @code{invert} is used, the grid color is the same as the
5026 video with inverted luma.
5029 The expression which sets the thickness of the grid line. Default value is @code{1}.
5031 See below for the list of accepted constants.
5034 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
5035 following constants:
5039 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
5043 horizontal and vertical chroma subsample values. For example for the
5044 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5048 The input grid cell width and height.
5051 The input sample aspect ratio.
5055 The x and y coordinates of some point of grid intersection (meant to configure offset).
5059 The width and height of the drawn cell.
5062 The thickness of the drawn cell.
5064 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
5065 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
5069 @subsection Examples
5073 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
5075 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
5079 Draw a white 3x3 grid with an opacity of 50%:
5081 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
5088 Draw a text string or text from a specified file on top of a video, using the
5089 libfreetype library.
5091 To enable compilation of this filter, you need to configure FFmpeg with
5092 @code{--enable-libfreetype}.
5093 To enable default font fallback and the @var{font} option you need to
5094 configure FFmpeg with @code{--enable-libfontconfig}.
5095 To enable the @var{text_shaping} option, you need to configure FFmpeg with
5096 @code{--enable-libfribidi}.
5100 It accepts the following parameters:
5105 Used to draw a box around text using the background color.
5106 The value must be either 1 (enable) or 0 (disable).
5107 The default value of @var{box} is 0.
5110 Set the width of the border to be drawn around the box using @var{boxcolor}.
5111 The default value of @var{boxborderw} is 0.
5114 The color to be used for drawing box around text. For the syntax of this
5115 option, check the "Color" section in the ffmpeg-utils manual.
5117 The default value of @var{boxcolor} is "white".
5120 Set the width of the border to be drawn around the text using @var{bordercolor}.
5121 The default value of @var{borderw} is 0.
5124 Set the color to be used for drawing border around text. For the syntax of this
5125 option, check the "Color" section in the ffmpeg-utils manual.
5127 The default value of @var{bordercolor} is "black".
5130 Select how the @var{text} is expanded. Can be either @code{none},
5131 @code{strftime} (deprecated) or
5132 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
5136 If true, check and fix text coords to avoid clipping.
5139 The color to be used for drawing fonts. For the syntax of this option, check
5140 the "Color" section in the ffmpeg-utils manual.
5142 The default value of @var{fontcolor} is "black".
5144 @item fontcolor_expr
5145 String which is expanded the same way as @var{text} to obtain dynamic
5146 @var{fontcolor} value. By default this option has empty value and is not
5147 processed. When this option is set, it overrides @var{fontcolor} option.
5150 The font family to be used for drawing text. By default Sans.
5153 The font file to be used for drawing text. The path must be included.
5154 This parameter is mandatory if the fontconfig support is disabled.
5157 This option does not exist, please see the timeline system
5160 Draw the text applying alpha blending. The value can
5161 be either a number between 0.0 and 1.0
5162 The expression accepts the same variables @var{x, y} do.
5163 The default value is 1.
5164 Please see fontcolor_expr
5167 The font size to be used for drawing text.
5168 The default value of @var{fontsize} is 16.
5171 If set to 1, attempt to shape the text (for example, reverse the order of
5172 right-to-left text and join Arabic characters) before drawing it.
5173 Otherwise, just draw the text exactly as given.
5174 By default 1 (if supported).
5177 The flags to be used for loading the fonts.
5179 The flags map the corresponding flags supported by libfreetype, and are
5180 a combination of the following values:
5187 @item vertical_layout
5188 @item force_autohint
5191 @item ignore_global_advance_width
5193 @item ignore_transform
5199 Default value is "default".
5201 For more information consult the documentation for the FT_LOAD_*
5205 The color to be used for drawing a shadow behind the drawn text. For the
5206 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
5208 The default value of @var{shadowcolor} is "black".
5212 The x and y offsets for the text shadow position with respect to the
5213 position of the text. They can be either positive or negative
5214 values. The default value for both is "0".
5217 The starting frame number for the n/frame_num variable. The default value
5221 The size in number of spaces to use for rendering the tab.
5225 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
5226 format. It can be used with or without text parameter. @var{timecode_rate}
5227 option must be specified.
5229 @item timecode_rate, rate, r
5230 Set the timecode frame rate (timecode only).
5233 The text string to be drawn. The text must be a sequence of UTF-8
5235 This parameter is mandatory if no file is specified with the parameter
5239 A text file containing text to be drawn. The text must be a sequence
5240 of UTF-8 encoded characters.
5242 This parameter is mandatory if no text string is specified with the
5243 parameter @var{text}.
5245 If both @var{text} and @var{textfile} are specified, an error is thrown.
5248 If set to 1, the @var{textfile} will be reloaded before each frame.
5249 Be sure to update it atomically, or it may be read partially, or even fail.
5253 The expressions which specify the offsets where text will be drawn
5254 within the video frame. They are relative to the top/left border of the
5257 The default value of @var{x} and @var{y} is "0".
5259 See below for the list of accepted constants and functions.
5262 The parameters for @var{x} and @var{y} are expressions containing the
5263 following constants and functions:
5267 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
5271 horizontal and vertical chroma subsample values. For example for the
5272 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5275 the height of each text line
5283 @item max_glyph_a, ascent
5284 the maximum distance from the baseline to the highest/upper grid
5285 coordinate used to place a glyph outline point, for all the rendered
5287 It is a positive value, due to the grid's orientation with the Y axis
5290 @item max_glyph_d, descent
5291 the maximum distance from the baseline to the lowest grid coordinate
5292 used to place a glyph outline point, for all the rendered glyphs.
5293 This is a negative value, due to the grid's orientation, with the Y axis
5297 maximum glyph height, that is the maximum height for all the glyphs
5298 contained in the rendered text, it is equivalent to @var{ascent} -
5302 maximum glyph width, that is the maximum width for all the glyphs
5303 contained in the rendered text
5306 the number of input frame, starting from 0
5308 @item rand(min, max)
5309 return a random number included between @var{min} and @var{max}
5312 The input sample aspect ratio.
5315 timestamp expressed in seconds, NAN if the input timestamp is unknown
5318 the height of the rendered text
5321 the width of the rendered text
5325 the x and y offset coordinates where the text is drawn.
5327 These parameters allow the @var{x} and @var{y} expressions to refer
5328 each other, so you can for example specify @code{y=x/dar}.
5331 @anchor{drawtext_expansion}
5332 @subsection Text expansion
5334 If @option{expansion} is set to @code{strftime},
5335 the filter recognizes strftime() sequences in the provided text and
5336 expands them accordingly. Check the documentation of strftime(). This
5337 feature is deprecated.
5339 If @option{expansion} is set to @code{none}, the text is printed verbatim.
5341 If @option{expansion} is set to @code{normal} (which is the default),
5342 the following expansion mechanism is used.
5344 The backslash character @samp{\}, followed by any character, always expands to
5345 the second character.
5347 Sequence of the form @code{%@{...@}} are expanded. The text between the
5348 braces is a function name, possibly followed by arguments separated by ':'.
5349 If the arguments contain special characters or delimiters (':' or '@}'),
5350 they should be escaped.
5352 Note that they probably must also be escaped as the value for the
5353 @option{text} option in the filter argument string and as the filter
5354 argument in the filtergraph description, and possibly also for the shell,
5355 that makes up to four levels of escaping; using a text file avoids these
5358 The following functions are available:
5363 The expression evaluation result.
5365 It must take one argument specifying the expression to be evaluated,
5366 which accepts the same constants and functions as the @var{x} and
5367 @var{y} values. Note that not all constants should be used, for
5368 example the text size is not known when evaluating the expression, so
5369 the constants @var{text_w} and @var{text_h} will have an undefined
5372 @item expr_int_format, eif
5373 Evaluate the expression's value and output as formatted integer.
5375 The first argument is the expression to be evaluated, just as for the @var{expr} function.
5376 The second argument specifies the output format. Allowed values are @samp{x},
5377 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
5378 @code{printf} function.
5379 The third parameter is optional and sets the number of positions taken by the output.
5380 It can be used to add padding with zeros from the left.
5383 The time at which the filter is running, expressed in UTC.
5384 It can accept an argument: a strftime() format string.
5387 The time at which the filter is running, expressed in the local time zone.
5388 It can accept an argument: a strftime() format string.
5391 Frame metadata. It must take one argument specifying metadata key.
5394 The frame number, starting from 0.
5397 A 1 character description of the current picture type.
5400 The timestamp of the current frame.
5401 It can take up to two arguments.
5403 The first argument is the format of the timestamp; it defaults to @code{flt}
5404 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
5405 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
5407 The second argument is an offset added to the timestamp.
5411 @subsection Examples
5415 Draw "Test Text" with font FreeSerif, using the default values for the
5416 optional parameters.
5419 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
5423 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
5424 and y=50 (counting from the top-left corner of the screen), text is
5425 yellow with a red box around it. Both the text and the box have an
5429 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
5430 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
5433 Note that the double quotes are not necessary if spaces are not used
5434 within the parameter list.
5437 Show the text at the center of the video frame:
5439 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
5443 Show a text line sliding from right to left in the last row of the video
5444 frame. The file @file{LONG_LINE} is assumed to contain a single line
5447 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
5451 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
5453 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
5457 Draw a single green letter "g", at the center of the input video.
5458 The glyph baseline is placed at half screen height.
5460 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
5464 Show text for 1 second every 3 seconds:
5466 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
5470 Use fontconfig to set the font. Note that the colons need to be escaped.
5472 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
5476 Print the date of a real-time encoding (see strftime(3)):
5478 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
5482 Show text fading in and out (appearing/disappearing):
5485 DS=1.0 # display start
5486 DE=10.0 # display end
5487 FID=1.5 # fade in duration
5488 FOD=5 # fade out duration
5489 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 @}"
5494 For more information about libfreetype, check:
5495 @url{http://www.freetype.org/}.
5497 For more information about fontconfig, check:
5498 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
5500 For more information about libfribidi, check:
5501 @url{http://fribidi.org/}.
5505 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
5507 The filter accepts the following options:
5512 Set low and high threshold values used by the Canny thresholding
5515 The high threshold selects the "strong" edge pixels, which are then
5516 connected through 8-connectivity with the "weak" edge pixels selected
5517 by the low threshold.
5519 @var{low} and @var{high} threshold values must be chosen in the range
5520 [0,1], and @var{low} should be lesser or equal to @var{high}.
5522 Default value for @var{low} is @code{20/255}, and default value for @var{high}
5526 Define the drawing mode.
5530 Draw white/gray wires on black background.
5533 Mix the colors to create a paint/cartoon effect.
5536 Default value is @var{wires}.
5539 @subsection Examples
5543 Standard edge detection with custom values for the hysteresis thresholding:
5545 edgedetect=low=0.1:high=0.4
5549 Painting effect without thresholding:
5551 edgedetect=mode=colormix:high=0
5556 Set brightness, contrast, saturation and approximate gamma adjustment.
5558 The filter accepts the following options:
5562 Set the contrast expression. The value must be a float value in range
5563 @code{-2.0} to @code{2.0}. The default value is "1".
5566 Set the brightness expression. The value must be a float value in
5567 range @code{-1.0} to @code{1.0}. The default value is "0".
5570 Set the saturation expression. The value must be a float in
5571 range @code{0.0} to @code{3.0}. The default value is "1".
5574 Set the gamma expression. The value must be a float in range
5575 @code{0.1} to @code{10.0}. The default value is "1".
5578 Set the gamma expression for red. The value must be a float in
5579 range @code{0.1} to @code{10.0}. The default value is "1".
5582 Set the gamma expression for green. The value must be a float in range
5583 @code{0.1} to @code{10.0}. The default value is "1".
5586 Set the gamma expression for blue. The value must be a float in range
5587 @code{0.1} to @code{10.0}. The default value is "1".
5590 Set the gamma weight expression. It can be used to reduce the effect
5591 of a high gamma value on bright image areas, e.g. keep them from
5592 getting overamplified and just plain white. The value must be a float
5593 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
5594 gamma correction all the way down while @code{1.0} leaves it at its
5595 full strength. Default is "1".
5598 Set when the expressions for brightness, contrast, saturation and
5599 gamma expressions are evaluated.
5601 It accepts the following values:
5604 only evaluate expressions once during the filter initialization or
5605 when a command is processed
5608 evaluate expressions for each incoming frame
5611 Default value is @samp{init}.
5614 The expressions accept the following parameters:
5617 frame count of the input frame starting from 0
5620 byte position of the corresponding packet in the input file, NAN if
5624 frame rate of the input video, NAN if the input frame rate is unknown
5627 timestamp expressed in seconds, NAN if the input timestamp is unknown
5630 @subsection Commands
5631 The filter supports the following commands:
5635 Set the contrast expression.
5638 Set the brightness expression.
5641 Set the saturation expression.
5644 Set the gamma expression.
5647 Set the gamma_r expression.
5650 Set gamma_g expression.
5653 Set gamma_b expression.
5656 Set gamma_weight expression.
5658 The command accepts the same syntax of the corresponding option.
5660 If the specified expression is not valid, it is kept at its current
5667 Apply erosion effect to the video.
5669 This filter replaces the pixel by the local(3x3) minimum.
5671 It accepts the following options:
5678 Limit the maximum change for each plane, default is 65535.
5679 If 0, plane will remain unchanged.
5682 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
5685 Flags to local 3x3 coordinates maps like this:
5692 @section extractplanes
5694 Extract color channel components from input video stream into
5695 separate grayscale video streams.
5697 The filter accepts the following option:
5701 Set plane(s) to extract.
5703 Available values for planes are:
5714 Choosing planes not available in the input will result in an error.
5715 That means you cannot select @code{r}, @code{g}, @code{b} planes
5716 with @code{y}, @code{u}, @code{v} planes at same time.
5719 @subsection Examples
5723 Extract luma, u and v color channel component from input video frame
5724 into 3 grayscale outputs:
5726 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
5732 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
5734 For each input image, the filter will compute the optimal mapping from
5735 the input to the output given the codebook length, that is the number
5736 of distinct output colors.
5738 This filter accepts the following options.
5741 @item codebook_length, l
5742 Set codebook length. The value must be a positive integer, and
5743 represents the number of distinct output colors. Default value is 256.
5746 Set the maximum number of iterations to apply for computing the optimal
5747 mapping. The higher the value the better the result and the higher the
5748 computation time. Default value is 1.
5751 Set a random seed, must be an integer included between 0 and
5752 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
5753 will try to use a good random seed on a best effort basis.
5756 Set pal8 output pixel format. This option does not work with codebook
5757 length greater than 256.
5762 Apply a fade-in/out effect to the input video.
5764 It accepts the following parameters:
5768 The effect type can be either "in" for a fade-in, or "out" for a fade-out
5770 Default is @code{in}.
5772 @item start_frame, s
5773 Specify the number of the frame to start applying the fade
5774 effect at. Default is 0.
5777 The number of frames that the fade effect lasts. At the end of the
5778 fade-in effect, the output video will have the same intensity as the input video.
5779 At the end of the fade-out transition, the output video will be filled with the
5780 selected @option{color}.
5784 If set to 1, fade only alpha channel, if one exists on the input.
5787 @item start_time, st
5788 Specify the timestamp (in seconds) of the frame to start to apply the fade
5789 effect. If both start_frame and start_time are specified, the fade will start at
5790 whichever comes last. Default is 0.
5793 The number of seconds for which the fade effect has to last. At the end of the
5794 fade-in effect the output video will have the same intensity as the input video,
5795 at the end of the fade-out transition the output video will be filled with the
5796 selected @option{color}.
5797 If both duration and nb_frames are specified, duration is used. Default is 0
5798 (nb_frames is used by default).
5801 Specify the color of the fade. Default is "black".
5804 @subsection Examples
5808 Fade in the first 30 frames of video:
5813 The command above is equivalent to:
5819 Fade out the last 45 frames of a 200-frame video:
5822 fade=type=out:start_frame=155:nb_frames=45
5826 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
5828 fade=in:0:25, fade=out:975:25
5832 Make the first 5 frames yellow, then fade in from frame 5-24:
5834 fade=in:5:20:color=yellow
5838 Fade in alpha over first 25 frames of video:
5840 fade=in:0:25:alpha=1
5844 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
5846 fade=t=in:st=5.5:d=0.5
5852 Apply arbitrary expressions to samples in frequency domain
5856 Adjust the dc value (gain) of the luma plane of the image. The filter
5857 accepts an integer value in range @code{0} to @code{1000}. The default
5858 value is set to @code{0}.
5861 Adjust the dc value (gain) of the 1st chroma plane of the image. The
5862 filter accepts an integer value in range @code{0} to @code{1000}. The
5863 default value is set to @code{0}.
5866 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
5867 filter accepts an integer value in range @code{0} to @code{1000}. The
5868 default value is set to @code{0}.
5871 Set the frequency domain weight expression for the luma plane.
5874 Set the frequency domain weight expression for the 1st chroma plane.
5877 Set the frequency domain weight expression for the 2nd chroma plane.
5879 The filter accepts the following variables:
5882 The coordinates of the current sample.
5886 The width and height of the image.
5889 @subsection Examples
5895 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
5901 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
5907 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
5914 Extract a single field from an interlaced image using stride
5915 arithmetic to avoid wasting CPU time. The output frames are marked as
5918 The filter accepts the following options:
5922 Specify whether to extract the top (if the value is @code{0} or
5923 @code{top}) or the bottom field (if the value is @code{1} or
5929 Field matching filter for inverse telecine. It is meant to reconstruct the
5930 progressive frames from a telecined stream. The filter does not drop duplicated
5931 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
5932 followed by a decimation filter such as @ref{decimate} in the filtergraph.
5934 The separation of the field matching and the decimation is notably motivated by
5935 the possibility of inserting a de-interlacing filter fallback between the two.
5936 If the source has mixed telecined and real interlaced content,
5937 @code{fieldmatch} will not be able to match fields for the interlaced parts.
5938 But these remaining combed frames will be marked as interlaced, and thus can be
5939 de-interlaced by a later filter such as @ref{yadif} before decimation.
5941 In addition to the various configuration options, @code{fieldmatch} can take an
5942 optional second stream, activated through the @option{ppsrc} option. If
5943 enabled, the frames reconstruction will be based on the fields and frames from
5944 this second stream. This allows the first input to be pre-processed in order to
5945 help the various algorithms of the filter, while keeping the output lossless
5946 (assuming the fields are matched properly). Typically, a field-aware denoiser,
5947 or brightness/contrast adjustments can help.
5949 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
5950 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
5951 which @code{fieldmatch} is based on. While the semantic and usage are very
5952 close, some behaviour and options names can differ.
5954 The @ref{decimate} filter currently only works for constant frame rate input.
5955 If your input has mixed telecined (30fps) and progressive content with a lower
5956 framerate like 24fps use the following filterchain to produce the necessary cfr
5957 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
5959 The filter accepts the following options:
5963 Specify the assumed field order of the input stream. Available values are:
5967 Auto detect parity (use FFmpeg's internal parity value).
5969 Assume bottom field first.
5971 Assume top field first.
5974 Note that it is sometimes recommended not to trust the parity announced by the
5977 Default value is @var{auto}.
5980 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
5981 sense that it won't risk creating jerkiness due to duplicate frames when
5982 possible, but if there are bad edits or blended fields it will end up
5983 outputting combed frames when a good match might actually exist. On the other
5984 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
5985 but will almost always find a good frame if there is one. The other values are
5986 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
5987 jerkiness and creating duplicate frames versus finding good matches in sections
5988 with bad edits, orphaned fields, blended fields, etc.
5990 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
5992 Available values are:
5996 2-way matching (p/c)
5998 2-way matching, and trying 3rd match if still combed (p/c + n)
6000 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
6002 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
6003 still combed (p/c + n + u/b)
6005 3-way matching (p/c/n)
6007 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
6008 detected as combed (p/c/n + u/b)
6011 The parenthesis at the end indicate the matches that would be used for that
6012 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
6015 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
6018 Default value is @var{pc_n}.
6021 Mark the main input stream as a pre-processed input, and enable the secondary
6022 input stream as the clean source to pick the fields from. See the filter
6023 introduction for more details. It is similar to the @option{clip2} feature from
6026 Default value is @code{0} (disabled).
6029 Set the field to match from. It is recommended to set this to the same value as
6030 @option{order} unless you experience matching failures with that setting. In
6031 certain circumstances changing the field that is used to match from can have a
6032 large impact on matching performance. Available values are:
6036 Automatic (same value as @option{order}).
6038 Match from the bottom field.
6040 Match from the top field.
6043 Default value is @var{auto}.
6046 Set whether or not chroma is included during the match comparisons. In most
6047 cases it is recommended to leave this enabled. You should set this to @code{0}
6048 only if your clip has bad chroma problems such as heavy rainbowing or other
6049 artifacts. Setting this to @code{0} could also be used to speed things up at
6050 the cost of some accuracy.
6052 Default value is @code{1}.
6056 These define an exclusion band which excludes the lines between @option{y0} and
6057 @option{y1} from being included in the field matching decision. An exclusion
6058 band can be used to ignore subtitles, a logo, or other things that may
6059 interfere with the matching. @option{y0} sets the starting scan line and
6060 @option{y1} sets the ending line; all lines in between @option{y0} and
6061 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
6062 @option{y0} and @option{y1} to the same value will disable the feature.
6063 @option{y0} and @option{y1} defaults to @code{0}.
6066 Set the scene change detection threshold as a percentage of maximum change on
6067 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
6068 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
6069 @option{scthresh} is @code{[0.0, 100.0]}.
6071 Default value is @code{12.0}.
6074 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
6075 account the combed scores of matches when deciding what match to use as the
6076 final match. Available values are:
6080 No final matching based on combed scores.
6082 Combed scores are only used when a scene change is detected.
6084 Use combed scores all the time.
6087 Default is @var{sc}.
6090 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
6091 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
6092 Available values are:
6096 No forced calculation.
6098 Force p/c/n calculations.
6100 Force p/c/n/u/b calculations.
6103 Default value is @var{none}.
6106 This is the area combing threshold used for combed frame detection. This
6107 essentially controls how "strong" or "visible" combing must be to be detected.
6108 Larger values mean combing must be more visible and smaller values mean combing
6109 can be less visible or strong and still be detected. Valid settings are from
6110 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
6111 be detected as combed). This is basically a pixel difference value. A good
6112 range is @code{[8, 12]}.
6114 Default value is @code{9}.
6117 Sets whether or not chroma is considered in the combed frame decision. Only
6118 disable this if your source has chroma problems (rainbowing, etc.) that are
6119 causing problems for the combed frame detection with chroma enabled. Actually,
6120 using @option{chroma}=@var{0} is usually more reliable, except for the case
6121 where there is chroma only combing in the source.
6123 Default value is @code{0}.
6127 Respectively set the x-axis and y-axis size of the window used during combed
6128 frame detection. This has to do with the size of the area in which
6129 @option{combpel} pixels are required to be detected as combed for a frame to be
6130 declared combed. See the @option{combpel} parameter description for more info.
6131 Possible values are any number that is a power of 2 starting at 4 and going up
6134 Default value is @code{16}.
6137 The number of combed pixels inside any of the @option{blocky} by
6138 @option{blockx} size blocks on the frame for the frame to be detected as
6139 combed. While @option{cthresh} controls how "visible" the combing must be, this
6140 setting controls "how much" combing there must be in any localized area (a
6141 window defined by the @option{blockx} and @option{blocky} settings) on the
6142 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
6143 which point no frames will ever be detected as combed). This setting is known
6144 as @option{MI} in TFM/VFM vocabulary.
6146 Default value is @code{80}.
6149 @anchor{p/c/n/u/b meaning}
6150 @subsection p/c/n/u/b meaning
6152 @subsubsection p/c/n
6154 We assume the following telecined stream:
6157 Top fields: 1 2 2 3 4
6158 Bottom fields: 1 2 3 4 4
6161 The numbers correspond to the progressive frame the fields relate to. Here, the
6162 first two frames are progressive, the 3rd and 4th are combed, and so on.
6164 When @code{fieldmatch} is configured to run a matching from bottom
6165 (@option{field}=@var{bottom}) this is how this input stream get transformed:
6170 B 1 2 3 4 4 <-- matching reference
6179 As a result of the field matching, we can see that some frames get duplicated.
6180 To perform a complete inverse telecine, you need to rely on a decimation filter
6181 after this operation. See for instance the @ref{decimate} filter.
6183 The same operation now matching from top fields (@option{field}=@var{top})
6188 T 1 2 2 3 4 <-- matching reference
6198 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
6199 basically, they refer to the frame and field of the opposite parity:
6202 @item @var{p} matches the field of the opposite parity in the previous frame
6203 @item @var{c} matches the field of the opposite parity in the current frame
6204 @item @var{n} matches the field of the opposite parity in the next frame
6209 The @var{u} and @var{b} matching are a bit special in the sense that they match
6210 from the opposite parity flag. In the following examples, we assume that we are
6211 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
6212 'x' is placed above and below each matched fields.
6214 With bottom matching (@option{field}=@var{bottom}):
6219 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6220 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6228 With top matching (@option{field}=@var{top}):
6233 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6234 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6242 @subsection Examples
6244 Simple IVTC of a top field first telecined stream:
6246 fieldmatch=order=tff:combmatch=none, decimate
6249 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
6251 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
6256 Transform the field order of the input video.
6258 It accepts the following parameters:
6263 The output field order. Valid values are @var{tff} for top field first or @var{bff}
6264 for bottom field first.
6267 The default value is @samp{tff}.
6269 The transformation is done by shifting the picture content up or down
6270 by one line, and filling the remaining line with appropriate picture content.
6271 This method is consistent with most broadcast field order converters.
6273 If the input video is not flagged as being interlaced, or it is already
6274 flagged as being of the required output field order, then this filter does
6275 not alter the incoming video.
6277 It is very useful when converting to or from PAL DV material,
6278 which is bottom field first.
6282 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
6287 Buffer input images and send them when they are requested.
6289 It is mainly useful when auto-inserted by the libavfilter
6292 It does not take parameters.
6296 Find a rectangular object
6298 It accepts the following options:
6302 Filepath of the object image, needs to be in gray8.
6305 Detection threshold, default is 0.5.
6308 Number of mipmaps, default is 3.
6310 @item xmin, ymin, xmax, ymax
6311 Specifies the rectangle in which to search.
6314 @subsection Examples
6318 Generate a representative palette of a given video using @command{ffmpeg}:
6320 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6326 Cover a rectangular object
6328 It accepts the following options:
6332 Filepath of the optional cover image, needs to be in yuv420.
6337 It accepts the following values:
6340 cover it by the supplied image
6342 cover it by interpolating the surrounding pixels
6345 Default value is @var{blur}.
6348 @subsection Examples
6352 Generate a representative palette of a given video using @command{ffmpeg}:
6354 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6361 Convert the input video to one of the specified pixel formats.
6362 Libavfilter will try to pick one that is suitable as input to
6365 It accepts the following parameters:
6369 A '|'-separated list of pixel format names, such as
6370 "pix_fmts=yuv420p|monow|rgb24".
6374 @subsection Examples
6378 Convert the input video to the @var{yuv420p} format
6380 format=pix_fmts=yuv420p
6383 Convert the input video to any of the formats in the list
6385 format=pix_fmts=yuv420p|yuv444p|yuv410p
6392 Convert the video to specified constant frame rate by duplicating or dropping
6393 frames as necessary.
6395 It accepts the following parameters:
6399 The desired output frame rate. The default is @code{25}.
6404 Possible values are:
6407 zero round towards 0
6411 round towards -infinity
6413 round towards +infinity
6417 The default is @code{near}.
6420 Assume the first PTS should be the given value, in seconds. This allows for
6421 padding/trimming at the start of stream. By default, no assumption is made
6422 about the first frame's expected PTS, so no padding or trimming is done.
6423 For example, this could be set to 0 to pad the beginning with duplicates of
6424 the first frame if a video stream starts after the audio stream or to trim any
6425 frames with a negative PTS.
6429 Alternatively, the options can be specified as a flat string:
6430 @var{fps}[:@var{round}].
6432 See also the @ref{setpts} filter.
6434 @subsection Examples
6438 A typical usage in order to set the fps to 25:
6444 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
6446 fps=fps=film:round=near
6452 Pack two different video streams into a stereoscopic video, setting proper
6453 metadata on supported codecs. The two views should have the same size and
6454 framerate and processing will stop when the shorter video ends. Please note
6455 that you may conveniently adjust view properties with the @ref{scale} and
6458 It accepts the following parameters:
6462 The desired packing format. Supported values are:
6467 The views are next to each other (default).
6470 The views are on top of each other.
6473 The views are packed by line.
6476 The views are packed by column.
6479 The views are temporally interleaved.
6488 # Convert left and right views into a frame-sequential video
6489 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
6491 # Convert views into a side-by-side video with the same output resolution as the input
6492 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
6497 Change the frame rate by interpolating new video output frames from the source
6500 This filter is not designed to function correctly with interlaced media. If
6501 you wish to change the frame rate of interlaced media then you are required
6502 to deinterlace before this filter and re-interlace after this filter.
6504 A description of the accepted options follows.
6508 Specify the output frames per second. This option can also be specified
6509 as a value alone. The default is @code{50}.
6512 Specify the start of a range where the output frame will be created as a
6513 linear interpolation of two frames. The range is [@code{0}-@code{255}],
6514 the default is @code{15}.
6517 Specify the end of a range where the output frame will be created as a
6518 linear interpolation of two frames. The range is [@code{0}-@code{255}],
6519 the default is @code{240}.
6522 Specify the level at which a scene change is detected as a value between
6523 0 and 100 to indicate a new scene; a low value reflects a low
6524 probability for the current frame to introduce a new scene, while a higher
6525 value means the current frame is more likely to be one.
6526 The default is @code{7}.
6529 Specify flags influencing the filter process.
6531 Available value for @var{flags} is:
6534 @item scene_change_detect, scd
6535 Enable scene change detection using the value of the option @var{scene}.
6536 This flag is enabled by default.
6542 Select one frame every N-th frame.
6544 This filter accepts the following option:
6547 Select frame after every @code{step} frames.
6548 Allowed values are positive integers higher than 0. Default value is @code{1}.
6554 Apply a frei0r effect to the input video.
6556 To enable the compilation of this filter, you need to install the frei0r
6557 header and configure FFmpeg with @code{--enable-frei0r}.
6559 It accepts the following parameters:
6564 The name of the frei0r effect to load. If the environment variable
6565 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
6566 directories specified by the colon-separated list in @env{FREIOR_PATH}.
6567 Otherwise, the standard frei0r paths are searched, in this order:
6568 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
6569 @file{/usr/lib/frei0r-1/}.
6572 A '|'-separated list of parameters to pass to the frei0r effect.
6576 A frei0r effect parameter can be a boolean (its value is either
6577 "y" or "n"), a double, a color (specified as
6578 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
6579 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
6580 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
6581 @var{X} and @var{Y} are floating point numbers) and/or a string.
6583 The number and types of parameters depend on the loaded effect. If an
6584 effect parameter is not specified, the default value is set.
6586 @subsection Examples
6590 Apply the distort0r effect, setting the first two double parameters:
6592 frei0r=filter_name=distort0r:filter_params=0.5|0.01
6596 Apply the colordistance effect, taking a color as the first parameter:
6598 frei0r=colordistance:0.2/0.3/0.4
6599 frei0r=colordistance:violet
6600 frei0r=colordistance:0x112233
6604 Apply the perspective effect, specifying the top left and top right image
6607 frei0r=perspective:0.2/0.2|0.8/0.2
6611 For more information, see
6612 @url{http://frei0r.dyne.org}
6616 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
6618 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
6619 processing filter, one of them is performed once per block, not per pixel.
6620 This allows for much higher speed.
6622 The filter accepts the following options:
6626 Set quality. This option defines the number of levels for averaging. It accepts
6627 an integer in the range 4-5. Default value is @code{4}.
6630 Force a constant quantization parameter. It accepts an integer in range 0-63.
6631 If not set, the filter will use the QP from the video stream (if available).
6634 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
6635 more details but also more artifacts, while higher values make the image smoother
6636 but also blurrier. Default value is @code{0} − PSNR optimal.
6639 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
6640 option may cause flicker since the B-Frames have often larger QP. Default is
6641 @code{0} (not enabled).
6647 The filter accepts the following options:
6651 Set the luminance expression.
6653 Set the chrominance blue expression.
6655 Set the chrominance red expression.
6657 Set the alpha expression.
6659 Set the red expression.
6661 Set the green expression.
6663 Set the blue expression.
6666 The colorspace is selected according to the specified options. If one
6667 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
6668 options is specified, the filter will automatically select a YCbCr
6669 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
6670 @option{blue_expr} options is specified, it will select an RGB
6673 If one of the chrominance expression is not defined, it falls back on the other
6674 one. If no alpha expression is specified it will evaluate to opaque value.
6675 If none of chrominance expressions are specified, they will evaluate
6676 to the luminance expression.
6678 The expressions can use the following variables and functions:
6682 The sequential number of the filtered frame, starting from @code{0}.
6686 The coordinates of the current sample.
6690 The width and height of the image.
6694 Width and height scale depending on the currently filtered plane. It is the
6695 ratio between the corresponding luma plane number of pixels and the current
6696 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
6697 @code{0.5,0.5} for chroma planes.
6700 Time of the current frame, expressed in seconds.
6703 Return the value of the pixel at location (@var{x},@var{y}) of the current
6707 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
6711 Return the value of the pixel at location (@var{x},@var{y}) of the
6712 blue-difference chroma plane. Return 0 if there is no such plane.
6715 Return the value of the pixel at location (@var{x},@var{y}) of the
6716 red-difference chroma plane. Return 0 if there is no such plane.
6721 Return the value of the pixel at location (@var{x},@var{y}) of the
6722 red/green/blue component. Return 0 if there is no such component.
6725 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
6726 plane. Return 0 if there is no such plane.
6729 For functions, if @var{x} and @var{y} are outside the area, the value will be
6730 automatically clipped to the closer edge.
6732 @subsection Examples
6736 Flip the image horizontally:
6742 Generate a bidimensional sine wave, with angle @code{PI/3} and a
6743 wavelength of 100 pixels:
6745 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
6749 Generate a fancy enigmatic moving light:
6751 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
6755 Generate a quick emboss effect:
6757 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
6761 Modify RGB components depending on pixel position:
6763 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
6767 Create a radial gradient that is the same size as the input (also see
6768 the @ref{vignette} filter):
6770 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
6774 Create a linear gradient to use as a mask for another filter, then
6775 compose with @ref{overlay}. In this example the video will gradually
6776 become more blurry from the top to the bottom of the y-axis as defined
6777 by the linear gradient:
6779 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
6785 Fix the banding artifacts that are sometimes introduced into nearly flat
6786 regions by truncation to 8bit color depth.
6787 Interpolate the gradients that should go where the bands are, and
6790 It is designed for playback only. Do not use it prior to
6791 lossy compression, because compression tends to lose the dither and
6792 bring back the bands.
6794 It accepts the following parameters:
6799 The maximum amount by which the filter will change any one pixel. This is also
6800 the threshold for detecting nearly flat regions. Acceptable values range from
6801 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
6805 The neighborhood to fit the gradient to. A larger radius makes for smoother
6806 gradients, but also prevents the filter from modifying the pixels near detailed
6807 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
6808 values will be clipped to the valid range.
6812 Alternatively, the options can be specified as a flat string:
6813 @var{strength}[:@var{radius}]
6815 @subsection Examples
6819 Apply the filter with a @code{3.5} strength and radius of @code{8}:
6825 Specify radius, omitting the strength (which will fall-back to the default
6836 Apply a Hald CLUT to a video stream.
6838 First input is the video stream to process, and second one is the Hald CLUT.
6839 The Hald CLUT input can be a simple picture or a complete video stream.
6841 The filter accepts the following options:
6845 Force termination when the shortest input terminates. Default is @code{0}.
6847 Continue applying the last CLUT after the end of the stream. A value of
6848 @code{0} disable the filter after the last frame of the CLUT is reached.
6849 Default is @code{1}.
6852 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
6853 filters share the same internals).
6855 More information about the Hald CLUT can be found on Eskil Steenberg's website
6856 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
6858 @subsection Workflow examples
6860 @subsubsection Hald CLUT video stream
6862 Generate an identity Hald CLUT stream altered with various effects:
6864 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
6867 Note: make sure you use a lossless codec.
6869 Then use it with @code{haldclut} to apply it on some random stream:
6871 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
6874 The Hald CLUT will be applied to the 10 first seconds (duration of
6875 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
6876 to the remaining frames of the @code{mandelbrot} stream.
6878 @subsubsection Hald CLUT with preview
6880 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
6881 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
6882 biggest possible square starting at the top left of the picture. The remaining
6883 padding pixels (bottom or right) will be ignored. This area can be used to add
6884 a preview of the Hald CLUT.
6886 Typically, the following generated Hald CLUT will be supported by the
6887 @code{haldclut} filter:
6890 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
6891 pad=iw+320 [padded_clut];
6892 smptebars=s=320x256, split [a][b];
6893 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
6894 [main][b] overlay=W-320" -frames:v 1 clut.png
6897 It contains the original and a preview of the effect of the CLUT: SMPTE color
6898 bars are displayed on the right-top, and below the same color bars processed by
6901 Then, the effect of this Hald CLUT can be visualized with:
6903 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
6908 Flip the input video horizontally.
6910 For example, to horizontally flip the input video with @command{ffmpeg}:
6912 ffmpeg -i in.avi -vf "hflip" out.avi
6916 This filter applies a global color histogram equalization on a
6919 It can be used to correct video that has a compressed range of pixel
6920 intensities. The filter redistributes the pixel intensities to
6921 equalize their distribution across the intensity range. It may be
6922 viewed as an "automatically adjusting contrast filter". This filter is
6923 useful only for correcting degraded or poorly captured source
6926 The filter accepts the following options:
6930 Determine the amount of equalization to be applied. As the strength
6931 is reduced, the distribution of pixel intensities more-and-more
6932 approaches that of the input frame. The value must be a float number
6933 in the range [0,1] and defaults to 0.200.
6936 Set the maximum intensity that can generated and scale the output
6937 values appropriately. The strength should be set as desired and then
6938 the intensity can be limited if needed to avoid washing-out. The value
6939 must be a float number in the range [0,1] and defaults to 0.210.
6942 Set the antibanding level. If enabled the filter will randomly vary
6943 the luminance of output pixels by a small amount to avoid banding of
6944 the histogram. Possible values are @code{none}, @code{weak} or
6945 @code{strong}. It defaults to @code{none}.
6950 Compute and draw a color distribution histogram for the input video.
6952 The computed histogram is a representation of the color component
6953 distribution in an image.
6955 The filter accepts the following options:
6961 It accepts the following values:
6964 Standard histogram that displays the color components distribution in an
6965 image. Displays color graph for each color component. Shows distribution of
6966 the Y, U, V, A or R, G, B components, depending on input format, in the
6967 current frame. Below each graph a color component scale meter is shown.
6970 Displays chroma values (U/V color placement) in a two dimensional
6971 graph (which is called a vectorscope). The brighter a pixel in the
6972 vectorscope, the more pixels of the input frame correspond to that pixel
6973 (i.e., more pixels have this chroma value). The V component is displayed on
6974 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
6975 side being V = 255. The U component is displayed on the vertical (Y) axis,
6976 with the top representing U = 0 and the bottom representing U = 255.
6978 The position of a white pixel in the graph corresponds to the chroma value of
6979 a pixel of the input clip. The graph can therefore be used to read the hue
6980 (color flavor) and the saturation (the dominance of the hue in the color). As
6981 the hue of a color changes, it moves around the square. At the center of the
6982 square the saturation is zero, which means that the corresponding pixel has no
6983 color. If the amount of a specific color is increased (while leaving the other
6984 colors unchanged) the saturation increases, and the indicator moves towards
6985 the edge of the square.
6988 Chroma values in vectorscope, similar as @code{color} but actual chroma values
6992 Per row/column color component graph. In row mode, the graph on the left side
6993 represents color component value 0 and the right side represents value = 255.
6994 In column mode, the top side represents color component value = 0 and bottom
6995 side represents value = 255.
6997 Default value is @code{levels}.
7000 Set height of level in @code{levels}. Default value is @code{200}.
7001 Allowed range is [50, 2048].
7004 Set height of color scale in @code{levels}. Default value is @code{12}.
7005 Allowed range is [0, 40].
7008 Set step for @code{waveform} mode. Smaller values are useful to find out how
7009 many values of the same luminance are distributed across input rows/columns.
7010 Default value is @code{10}. Allowed range is [1, 255].
7013 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
7014 Default is @code{row}.
7016 @item waveform_mirror
7017 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
7018 means mirrored. In mirrored mode, higher values will be represented on the left
7019 side for @code{row} mode and at the top for @code{column} mode. Default is
7020 @code{0} (unmirrored).
7023 Set display mode for @code{waveform} and @code{levels}.
7024 It accepts the following values:
7027 Display separate graph for the color components side by side in
7028 @code{row} waveform mode or one below the other in @code{column} waveform mode
7029 for @code{waveform} histogram mode. For @code{levels} histogram mode,
7030 per color component graphs are placed below each other.
7032 Using this display mode in @code{waveform} histogram mode makes it easy to
7033 spot color casts in the highlights and shadows of an image, by comparing the
7034 contours of the top and the bottom graphs of each waveform. Since whites,
7035 grays, and blacks are characterized by exactly equal amounts of red, green,
7036 and blue, neutral areas of the picture should display three waveforms of
7037 roughly equal width/height. If not, the correction is easy to perform by
7038 making level adjustments the three waveforms.
7041 Presents information identical to that in the @code{parade}, except
7042 that the graphs representing color components are superimposed directly
7045 This display mode in @code{waveform} histogram mode makes it easier to spot
7046 relative differences or similarities in overlapping areas of the color
7047 components that are supposed to be identical, such as neutral whites, grays,
7050 Default is @code{parade}.
7053 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
7054 Default is @code{linear}.
7057 Set what color components to display for mode @code{levels}.
7058 Default is @code{7}.
7061 @subsection Examples
7066 Calculate and draw histogram:
7068 ffplay -i input -vf histogram
7076 This is a high precision/quality 3d denoise filter. It aims to reduce
7077 image noise, producing smooth images and making still images really
7078 still. It should enhance compressibility.
7080 It accepts the following optional parameters:
7084 A non-negative floating point number which specifies spatial luma strength.
7087 @item chroma_spatial
7088 A non-negative floating point number which specifies spatial chroma strength.
7089 It defaults to 3.0*@var{luma_spatial}/4.0.
7092 A floating point number which specifies luma temporal strength. It defaults to
7093 6.0*@var{luma_spatial}/4.0.
7096 A floating point number which specifies chroma temporal strength. It defaults to
7097 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
7102 Apply a high-quality magnification filter designed for pixel art. This filter
7103 was originally created by Maxim Stepin.
7105 It accepts the following option:
7109 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
7110 @code{hq3x} and @code{4} for @code{hq4x}.
7111 Default is @code{3}.
7115 Stack input videos horizontally.
7117 All streams must be of same pixel format and of same height.
7119 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
7120 to create same output.
7122 The filter accept the following option:
7126 Set number of input streams. Default is 2.
7131 Modify the hue and/or the saturation of the input.
7133 It accepts the following parameters:
7137 Specify the hue angle as a number of degrees. It accepts an expression,
7138 and defaults to "0".
7141 Specify the saturation in the [-10,10] range. It accepts an expression and
7145 Specify the hue angle as a number of radians. It accepts an
7146 expression, and defaults to "0".
7149 Specify the brightness in the [-10,10] range. It accepts an expression and
7153 @option{h} and @option{H} are mutually exclusive, and can't be
7154 specified at the same time.
7156 The @option{b}, @option{h}, @option{H} and @option{s} option values are
7157 expressions containing the following constants:
7161 frame count of the input frame starting from 0
7164 presentation timestamp of the input frame expressed in time base units
7167 frame rate of the input video, NAN if the input frame rate is unknown
7170 timestamp expressed in seconds, NAN if the input timestamp is unknown
7173 time base of the input video
7176 @subsection Examples
7180 Set the hue to 90 degrees and the saturation to 1.0:
7186 Same command but expressing the hue in radians:
7192 Rotate hue and make the saturation swing between 0
7193 and 2 over a period of 1 second:
7195 hue="H=2*PI*t: s=sin(2*PI*t)+1"
7199 Apply a 3 seconds saturation fade-in effect starting at 0:
7204 The general fade-in expression can be written as:
7206 hue="s=min(0\, max((t-START)/DURATION\, 1))"
7210 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
7212 hue="s=max(0\, min(1\, (8-t)/3))"
7215 The general fade-out expression can be written as:
7217 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
7222 @subsection Commands
7224 This filter supports the following commands:
7230 Modify the hue and/or the saturation and/or brightness of the input video.
7231 The command accepts the same syntax of the corresponding option.
7233 If the specified expression is not valid, it is kept at its current
7239 Detect video interlacing type.
7241 This filter tries to detect if the input frames as interlaced, progressive,
7242 top or bottom field first. It will also try and detect fields that are
7243 repeated between adjacent frames (a sign of telecine).
7245 Single frame detection considers only immediately adjacent frames when classifying each frame.
7246 Multiple frame detection incorporates the classification history of previous frames.
7248 The filter will log these metadata values:
7251 @item single.current_frame
7252 Detected type of current frame using single-frame detection. One of:
7253 ``tff'' (top field first), ``bff'' (bottom field first),
7254 ``progressive'', or ``undetermined''
7257 Cumulative number of frames detected as top field first using single-frame detection.
7260 Cumulative number of frames detected as top field first using multiple-frame detection.
7263 Cumulative number of frames detected as bottom field first using single-frame detection.
7265 @item multiple.current_frame
7266 Detected type of current frame using multiple-frame detection. One of:
7267 ``tff'' (top field first), ``bff'' (bottom field first),
7268 ``progressive'', or ``undetermined''
7271 Cumulative number of frames detected as bottom field first using multiple-frame detection.
7273 @item single.progressive
7274 Cumulative number of frames detected as progressive using single-frame detection.
7276 @item multiple.progressive
7277 Cumulative number of frames detected as progressive using multiple-frame detection.
7279 @item single.undetermined
7280 Cumulative number of frames that could not be classified using single-frame detection.
7282 @item multiple.undetermined
7283 Cumulative number of frames that could not be classified using multiple-frame detection.
7285 @item repeated.current_frame
7286 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
7288 @item repeated.neither
7289 Cumulative number of frames with no repeated field.
7292 Cumulative number of frames with the top field repeated from the previous frame's top field.
7294 @item repeated.bottom
7295 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
7298 The filter accepts the following options:
7302 Set interlacing threshold.
7304 Set progressive threshold.
7306 Threshold for repeated field detection.
7308 Number of frames after which a given frame's contribution to the
7309 statistics is halved (i.e., it contributes only 0.5 to it's
7310 classification). The default of 0 means that all frames seen are given
7311 full weight of 1.0 forever.
7312 @item analyze_interlaced_flag
7313 When this is not 0 then idet will use the specified number of frames to determine
7314 if the interlaced flag is accurate, it will not count undetermined frames.
7315 If the flag is found to be accurate it will be used without any further
7316 computations, if it is found to be inaccurate it will be cleared without any
7317 further computations. This allows inserting the idet filter as a low computational
7318 method to clean up the interlaced flag
7323 Deinterleave or interleave fields.
7325 This filter allows one to process interlaced images fields without
7326 deinterlacing them. Deinterleaving splits the input frame into 2
7327 fields (so called half pictures). Odd lines are moved to the top
7328 half of the output image, even lines to the bottom half.
7329 You can process (filter) them independently and then re-interleave them.
7331 The filter accepts the following options:
7335 @item chroma_mode, c
7337 Available values for @var{luma_mode}, @var{chroma_mode} and
7338 @var{alpha_mode} are:
7344 @item deinterleave, d
7345 Deinterleave fields, placing one above the other.
7348 Interleave fields. Reverse the effect of deinterleaving.
7350 Default value is @code{none}.
7353 @item chroma_swap, cs
7354 @item alpha_swap, as
7355 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
7360 Apply inflate effect to the video.
7362 This filter replaces the pixel by the local(3x3) average by taking into account
7363 only values higher than the pixel.
7365 It accepts the following options:
7372 Limit the maximum change for each plane, default is 65535.
7373 If 0, plane will remain unchanged.
7378 Simple interlacing filter from progressive contents. This interleaves upper (or
7379 lower) lines from odd frames with lower (or upper) lines from even frames,
7380 halving the frame rate and preserving image height.
7383 Original Original New Frame
7384 Frame 'j' Frame 'j+1' (tff)
7385 ========== =========== ==================
7386 Line 0 --------------------> Frame 'j' Line 0
7387 Line 1 Line 1 ----> Frame 'j+1' Line 1
7388 Line 2 ---------------------> Frame 'j' Line 2
7389 Line 3 Line 3 ----> Frame 'j+1' Line 3
7391 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
7394 It accepts the following optional parameters:
7398 This determines whether the interlaced frame is taken from the even
7399 (tff - default) or odd (bff) lines of the progressive frame.
7402 Enable (default) or disable the vertical lowpass filter to avoid twitter
7403 interlacing and reduce moire patterns.
7408 Deinterlace input video by applying Donald Graft's adaptive kernel
7409 deinterling. Work on interlaced parts of a video to produce
7412 The description of the accepted parameters follows.
7416 Set the threshold which affects the filter's tolerance when
7417 determining if a pixel line must be processed. It must be an integer
7418 in the range [0,255] and defaults to 10. A value of 0 will result in
7419 applying the process on every pixels.
7422 Paint pixels exceeding the threshold value to white if set to 1.
7426 Set the fields order. Swap fields if set to 1, leave fields alone if
7430 Enable additional sharpening if set to 1. Default is 0.
7433 Enable twoway sharpening if set to 1. Default is 0.
7436 @subsection Examples
7440 Apply default values:
7442 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
7446 Enable additional sharpening:
7452 Paint processed pixels in white:
7458 @section lenscorrection
7460 Correct radial lens distortion
7462 This filter can be used to correct for radial distortion as can result from the use
7463 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
7464 one can use tools available for example as part of opencv or simply trial-and-error.
7465 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
7466 and extract the k1 and k2 coefficients from the resulting matrix.
7468 Note that effectively the same filter is available in the open-source tools Krita and
7469 Digikam from the KDE project.
7471 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
7472 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
7473 brightness distribution, so you may want to use both filters together in certain
7474 cases, though you will have to take care of ordering, i.e. whether vignetting should
7475 be applied before or after lens correction.
7479 The filter accepts the following options:
7483 Relative x-coordinate of the focal point of the image, and thereby the center of the
7484 distortion. This value has a range [0,1] and is expressed as fractions of the image
7487 Relative y-coordinate of the focal point of the image, and thereby the center of the
7488 distortion. This value has a range [0,1] and is expressed as fractions of the image
7491 Coefficient of the quadratic correction term. 0.5 means no correction.
7493 Coefficient of the double quadratic correction term. 0.5 means no correction.
7496 The formula that generates the correction is:
7498 @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)
7500 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
7501 distances from the focal point in the source and target images, respectively.
7506 Apply a 3D LUT to an input video.
7508 The filter accepts the following options:
7512 Set the 3D LUT file name.
7514 Currently supported formats:
7526 Select interpolation mode.
7528 Available values are:
7532 Use values from the nearest defined point.
7534 Interpolate values using the 8 points defining a cube.
7536 Interpolate values using a tetrahedron.
7540 @section lut, lutrgb, lutyuv
7542 Compute a look-up table for binding each pixel component input value
7543 to an output value, and apply it to the input video.
7545 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
7546 to an RGB input video.
7548 These filters accept the following parameters:
7551 set first pixel component expression
7553 set second pixel component expression
7555 set third pixel component expression
7557 set fourth pixel component expression, corresponds to the alpha component
7560 set red component expression
7562 set green component expression
7564 set blue component expression
7566 alpha component expression
7569 set Y/luminance component expression
7571 set U/Cb component expression
7573 set V/Cr component expression
7576 Each of them specifies the expression to use for computing the lookup table for
7577 the corresponding pixel component values.
7579 The exact component associated to each of the @var{c*} options depends on the
7582 The @var{lut} filter requires either YUV or RGB pixel formats in input,
7583 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
7585 The expressions can contain the following constants and functions:
7590 The input width and height.
7593 The input value for the pixel component.
7596 The input value, clipped to the @var{minval}-@var{maxval} range.
7599 The maximum value for the pixel component.
7602 The minimum value for the pixel component.
7605 The negated value for the pixel component value, clipped to the
7606 @var{minval}-@var{maxval} range; it corresponds to the expression
7607 "maxval-clipval+minval".
7610 The computed value in @var{val}, clipped to the
7611 @var{minval}-@var{maxval} range.
7613 @item gammaval(gamma)
7614 The computed gamma correction value of the pixel component value,
7615 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
7617 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
7621 All expressions default to "val".
7623 @subsection Examples
7629 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
7630 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
7633 The above is the same as:
7635 lutrgb="r=negval:g=negval:b=negval"
7636 lutyuv="y=negval:u=negval:v=negval"
7646 Remove chroma components, turning the video into a graytone image:
7648 lutyuv="u=128:v=128"
7652 Apply a luma burning effect:
7658 Remove green and blue components:
7664 Set a constant alpha channel value on input:
7666 format=rgba,lutrgb=a="maxval-minval/2"
7670 Correct luminance gamma by a factor of 0.5:
7672 lutyuv=y=gammaval(0.5)
7676 Discard least significant bits of luma:
7678 lutyuv=y='bitand(val, 128+64+32)'
7682 @section maskedmerge
7684 Merge the first input stream with the second input stream using per pixel
7685 weights in the third input stream.
7687 A value of 0 in the third stream pixel component means that pixel component
7688 from first stream is returned unchanged, while maximum value (eg. 255 for
7689 8-bit videos) means that pixel component from second stream is returned
7690 unchanged. Intermediate values define the amount of merging between both
7691 input stream's pixel components.
7693 This filter accepts the following options:
7696 Set which planes will be processed as bitmap, unprocessed planes will be
7697 copied from first stream.
7698 By default value 0xf, all planes will be processed.
7703 Apply motion-compensation deinterlacing.
7705 It needs one field per frame as input and must thus be used together
7706 with yadif=1/3 or equivalent.
7708 This filter accepts the following options:
7711 Set the deinterlacing mode.
7713 It accepts one of the following values:
7718 use iterative motion estimation
7720 like @samp{slow}, but use multiple reference frames.
7722 Default value is @samp{fast}.
7725 Set the picture field parity assumed for the input video. It must be
7726 one of the following values:
7730 assume top field first
7732 assume bottom field first
7735 Default value is @samp{bff}.
7738 Set per-block quantization parameter (QP) used by the internal
7741 Higher values should result in a smoother motion vector field but less
7742 optimal individual vectors. Default value is 1.
7745 @section mergeplanes
7747 Merge color channel components from several video streams.
7749 The filter accepts up to 4 input streams, and merge selected input
7750 planes to the output video.
7752 This filter accepts the following options:
7755 Set input to output plane mapping. Default is @code{0}.
7757 The mappings is specified as a bitmap. It should be specified as a
7758 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
7759 mapping for the first plane of the output stream. 'A' sets the number of
7760 the input stream to use (from 0 to 3), and 'a' the plane number of the
7761 corresponding input to use (from 0 to 3). The rest of the mappings is
7762 similar, 'Bb' describes the mapping for the output stream second
7763 plane, 'Cc' describes the mapping for the output stream third plane and
7764 'Dd' describes the mapping for the output stream fourth plane.
7767 Set output pixel format. Default is @code{yuva444p}.
7770 @subsection Examples
7774 Merge three gray video streams of same width and height into single video stream:
7776 [a0][a1][a2]mergeplanes=0x001020:yuv444p
7780 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
7782 [a0][a1]mergeplanes=0x00010210:yuva444p
7786 Swap Y and A plane in yuva444p stream:
7788 format=yuva444p,mergeplanes=0x03010200:yuva444p
7792 Swap U and V plane in yuv420p stream:
7794 format=yuv420p,mergeplanes=0x000201:yuv420p
7798 Cast a rgb24 clip to yuv444p:
7800 format=rgb24,mergeplanes=0x000102:yuv444p
7806 Drop frames that do not differ greatly from the previous frame in
7807 order to reduce frame rate.
7809 The main use of this filter is for very-low-bitrate encoding
7810 (e.g. streaming over dialup modem), but it could in theory be used for
7811 fixing movies that were inverse-telecined incorrectly.
7813 A description of the accepted options follows.
7817 Set the maximum number of consecutive frames which can be dropped (if
7818 positive), or the minimum interval between dropped frames (if
7819 negative). If the value is 0, the frame is dropped unregarding the
7820 number of previous sequentially dropped frames.
7827 Set the dropping threshold values.
7829 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
7830 represent actual pixel value differences, so a threshold of 64
7831 corresponds to 1 unit of difference for each pixel, or the same spread
7832 out differently over the block.
7834 A frame is a candidate for dropping if no 8x8 blocks differ by more
7835 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
7836 meaning the whole image) differ by more than a threshold of @option{lo}.
7838 Default value for @option{hi} is 64*12, default value for @option{lo} is
7839 64*5, and default value for @option{frac} is 0.33.
7847 It accepts an integer in input; if non-zero it negates the
7848 alpha component (if available). The default value in input is 0.
7852 Force libavfilter not to use any of the specified pixel formats for the
7853 input to the next filter.
7855 It accepts the following parameters:
7859 A '|'-separated list of pixel format names, such as
7860 apix_fmts=yuv420p|monow|rgb24".
7864 @subsection Examples
7868 Force libavfilter to use a format different from @var{yuv420p} for the
7869 input to the vflip filter:
7871 noformat=pix_fmts=yuv420p,vflip
7875 Convert the input video to any of the formats not contained in the list:
7877 noformat=yuv420p|yuv444p|yuv410p
7883 Add noise on video input frame.
7885 The filter accepts the following options:
7893 Set noise seed for specific pixel component or all pixel components in case
7894 of @var{all_seed}. Default value is @code{123457}.
7896 @item all_strength, alls
7897 @item c0_strength, c0s
7898 @item c1_strength, c1s
7899 @item c2_strength, c2s
7900 @item c3_strength, c3s
7901 Set noise strength for specific pixel component or all pixel components in case
7902 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
7904 @item all_flags, allf
7909 Set pixel component flags or set flags for all components if @var{all_flags}.
7910 Available values for component flags are:
7913 averaged temporal noise (smoother)
7915 mix random noise with a (semi)regular pattern
7917 temporal noise (noise pattern changes between frames)
7919 uniform noise (gaussian otherwise)
7923 @subsection Examples
7925 Add temporal and uniform noise to input video:
7927 noise=alls=20:allf=t+u
7932 Pass the video source unchanged to the output.
7935 Optical Character Recognition
7937 This filter uses Tesseract for optical character recognition.
7939 It accepts the following options:
7943 Set datapath to tesseract data. Default is to use whatever was
7944 set at installation.
7947 Set language, default is "eng".
7950 Set character whitelist.
7953 Set character blacklist.
7956 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
7960 Apply a video transform using libopencv.
7962 To enable this filter, install the libopencv library and headers and
7963 configure FFmpeg with @code{--enable-libopencv}.
7965 It accepts the following parameters:
7970 The name of the libopencv filter to apply.
7973 The parameters to pass to the libopencv filter. If not specified, the default
7978 Refer to the official libopencv documentation for more precise
7980 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
7982 Several libopencv filters are supported; see the following subsections.
7987 Dilate an image by using a specific structuring element.
7988 It corresponds to the libopencv function @code{cvDilate}.
7990 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
7992 @var{struct_el} represents a structuring element, and has the syntax:
7993 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
7995 @var{cols} and @var{rows} represent the number of columns and rows of
7996 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
7997 point, and @var{shape} the shape for the structuring element. @var{shape}
7998 must be "rect", "cross", "ellipse", or "custom".
8000 If the value for @var{shape} is "custom", it must be followed by a
8001 string of the form "=@var{filename}". The file with name
8002 @var{filename} is assumed to represent a binary image, with each
8003 printable character corresponding to a bright pixel. When a custom
8004 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
8005 or columns and rows of the read file are assumed instead.
8007 The default value for @var{struct_el} is "3x3+0x0/rect".
8009 @var{nb_iterations} specifies the number of times the transform is
8010 applied to the image, and defaults to 1.
8014 # Use the default values
8017 # Dilate using a structuring element with a 5x5 cross, iterating two times
8018 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
8020 # Read the shape from the file diamond.shape, iterating two times.
8021 # The file diamond.shape may contain a pattern of characters like this
8027 # The specified columns and rows are ignored
8028 # but the anchor point coordinates are not
8029 ocv=dilate:0x0+2x2/custom=diamond.shape|2
8034 Erode an image by using a specific structuring element.
8035 It corresponds to the libopencv function @code{cvErode}.
8037 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
8038 with the same syntax and semantics as the @ref{dilate} filter.
8042 Smooth the input video.
8044 The filter takes the following parameters:
8045 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
8047 @var{type} is the type of smooth filter to apply, and must be one of
8048 the following values: "blur", "blur_no_scale", "median", "gaussian",
8049 or "bilateral". The default value is "gaussian".
8051 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
8052 depend on the smooth type. @var{param1} and
8053 @var{param2} accept integer positive values or 0. @var{param3} and
8054 @var{param4} accept floating point values.
8056 The default value for @var{param1} is 3. The default value for the
8057 other parameters is 0.
8059 These parameters correspond to the parameters assigned to the
8060 libopencv function @code{cvSmooth}.
8065 Overlay one video on top of another.
8067 It takes two inputs and has one output. The first input is the "main"
8068 video on which the second input is overlaid.
8070 It accepts the following parameters:
8072 A description of the accepted options follows.
8077 Set the expression for the x and y coordinates of the overlaid video
8078 on the main video. Default value is "0" for both expressions. In case
8079 the expression is invalid, it is set to a huge value (meaning that the
8080 overlay will not be displayed within the output visible area).
8083 The action to take when EOF is encountered on the secondary input; it accepts
8084 one of the following values:
8088 Repeat the last frame (the default).
8092 Pass the main input through.
8096 Set when the expressions for @option{x}, and @option{y} are evaluated.
8098 It accepts the following values:
8101 only evaluate expressions once during the filter initialization or
8102 when a command is processed
8105 evaluate expressions for each incoming frame
8108 Default value is @samp{frame}.
8111 If set to 1, force the output to terminate when the shortest input
8112 terminates. Default value is 0.
8115 Set the format for the output video.
8117 It accepts the following values:
8132 Default value is @samp{yuv420}.
8134 @item rgb @emph{(deprecated)}
8135 If set to 1, force the filter to accept inputs in the RGB
8136 color space. Default value is 0. This option is deprecated, use
8137 @option{format} instead.
8140 If set to 1, force the filter to draw the last overlay frame over the
8141 main input until the end of the stream. A value of 0 disables this
8142 behavior. Default value is 1.
8145 The @option{x}, and @option{y} expressions can contain the following
8151 The main input width and height.
8155 The overlay input width and height.
8159 The computed values for @var{x} and @var{y}. They are evaluated for
8164 horizontal and vertical chroma subsample values of the output
8165 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
8169 the number of input frame, starting from 0
8172 the position in the file of the input frame, NAN if unknown
8175 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
8179 Note that the @var{n}, @var{pos}, @var{t} variables are available only
8180 when evaluation is done @emph{per frame}, and will evaluate to NAN
8181 when @option{eval} is set to @samp{init}.
8183 Be aware that frames are taken from each input video in timestamp
8184 order, hence, if their initial timestamps differ, it is a good idea
8185 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
8186 have them begin in the same zero timestamp, as the example for
8187 the @var{movie} filter does.
8189 You can chain together more overlays but you should test the
8190 efficiency of such approach.
8192 @subsection Commands
8194 This filter supports the following commands:
8198 Modify the x and y of the overlay input.
8199 The command accepts the same syntax of the corresponding option.
8201 If the specified expression is not valid, it is kept at its current
8205 @subsection Examples
8209 Draw the overlay at 10 pixels from the bottom right corner of the main
8212 overlay=main_w-overlay_w-10:main_h-overlay_h-10
8215 Using named options the example above becomes:
8217 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
8221 Insert a transparent PNG logo in the bottom left corner of the input,
8222 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
8224 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
8228 Insert 2 different transparent PNG logos (second logo on bottom
8229 right corner) using the @command{ffmpeg} tool:
8231 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
8235 Add a transparent color layer on top of the main video; @code{WxH}
8236 must specify the size of the main input to the overlay filter:
8238 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
8242 Play an original video and a filtered version (here with the deshake
8243 filter) side by side using the @command{ffplay} tool:
8245 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
8248 The above command is the same as:
8250 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
8254 Make a sliding overlay appearing from the left to the right top part of the
8255 screen starting since time 2:
8257 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
8261 Compose output by putting two input videos side to side:
8263 ffmpeg -i left.avi -i right.avi -filter_complex "
8264 nullsrc=size=200x100 [background];
8265 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
8266 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
8267 [background][left] overlay=shortest=1 [background+left];
8268 [background+left][right] overlay=shortest=1:x=100 [left+right]
8273 Mask 10-20 seconds of a video by applying the delogo filter to a section
8275 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
8276 -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]'
8281 Chain several overlays in cascade:
8283 nullsrc=s=200x200 [bg];
8284 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
8285 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
8286 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
8287 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
8288 [in3] null, [mid2] overlay=100:100 [out0]
8295 Apply Overcomplete Wavelet denoiser.
8297 The filter accepts the following options:
8303 Larger depth values will denoise lower frequency components more, but
8304 slow down filtering.
8306 Must be an int in the range 8-16, default is @code{8}.
8308 @item luma_strength, ls
8311 Must be a double value in the range 0-1000, default is @code{1.0}.
8313 @item chroma_strength, cs
8314 Set chroma strength.
8316 Must be a double value in the range 0-1000, default is @code{1.0}.
8322 Add paddings to the input image, and place the original input at the
8323 provided @var{x}, @var{y} coordinates.
8325 It accepts the following parameters:
8330 Specify an expression for the size of the output image with the
8331 paddings added. If the value for @var{width} or @var{height} is 0, the
8332 corresponding input size is used for the output.
8334 The @var{width} expression can reference the value set by the
8335 @var{height} expression, and vice versa.
8337 The default value of @var{width} and @var{height} is 0.
8341 Specify the offsets to place the input image at within the padded area,
8342 with respect to the top/left border of the output image.
8344 The @var{x} expression can reference the value set by the @var{y}
8345 expression, and vice versa.
8347 The default value of @var{x} and @var{y} is 0.
8350 Specify the color of the padded area. For the syntax of this option,
8351 check the "Color" section in the ffmpeg-utils manual.
8353 The default value of @var{color} is "black".
8356 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
8357 options are expressions containing the following constants:
8362 The input video width and height.
8366 These are the same as @var{in_w} and @var{in_h}.
8370 The output width and height (the size of the padded area), as
8371 specified by the @var{width} and @var{height} expressions.
8375 These are the same as @var{out_w} and @var{out_h}.
8379 The x and y offsets as specified by the @var{x} and @var{y}
8380 expressions, or NAN if not yet specified.
8383 same as @var{iw} / @var{ih}
8386 input sample aspect ratio
8389 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
8393 The horizontal and vertical chroma subsample values. For example for the
8394 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8397 @subsection Examples
8401 Add paddings with the color "violet" to the input video. The output video
8402 size is 640x480, and the top-left corner of the input video is placed at
8405 pad=640:480:0:40:violet
8408 The example above is equivalent to the following command:
8410 pad=width=640:height=480:x=0:y=40:color=violet
8414 Pad the input to get an output with dimensions increased by 3/2,
8415 and put the input video at the center of the padded area:
8417 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
8421 Pad the input to get a squared output with size equal to the maximum
8422 value between the input width and height, and put the input video at
8423 the center of the padded area:
8425 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
8429 Pad the input to get a final w/h ratio of 16:9:
8431 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
8435 In case of anamorphic video, in order to set the output display aspect
8436 correctly, it is necessary to use @var{sar} in the expression,
8437 according to the relation:
8439 (ih * X / ih) * sar = output_dar
8440 X = output_dar / sar
8443 Thus the previous example needs to be modified to:
8445 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
8449 Double the output size and put the input video in the bottom-right
8450 corner of the output padded area:
8452 pad="2*iw:2*ih:ow-iw:oh-ih"
8459 Generate one palette for a whole video stream.
8461 It accepts the following options:
8465 Set the maximum number of colors to quantize in the palette.
8466 Note: the palette will still contain 256 colors; the unused palette entries
8469 @item reserve_transparent
8470 Create a palette of 255 colors maximum and reserve the last one for
8471 transparency. Reserving the transparency color is useful for GIF optimization.
8472 If not set, the maximum of colors in the palette will be 256. You probably want
8473 to disable this option for a standalone image.
8477 Set statistics mode.
8479 It accepts the following values:
8482 Compute full frame histograms.
8484 Compute histograms only for the part that differs from previous frame. This
8485 might be relevant to give more importance to the moving part of your input if
8486 the background is static.
8489 Default value is @var{full}.
8492 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
8493 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
8494 color quantization of the palette. This information is also visible at
8495 @var{info} logging level.
8497 @subsection Examples
8501 Generate a representative palette of a given video using @command{ffmpeg}:
8503 ffmpeg -i input.mkv -vf palettegen palette.png
8509 Use a palette to downsample an input video stream.
8511 The filter takes two inputs: one video stream and a palette. The palette must
8512 be a 256 pixels image.
8514 It accepts the following options:
8518 Select dithering mode. Available algorithms are:
8521 Ordered 8x8 bayer dithering (deterministic)
8523 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
8524 Note: this dithering is sometimes considered "wrong" and is included as a
8526 @item floyd_steinberg
8527 Floyd and Steingberg dithering (error diffusion)
8529 Frankie Sierra dithering v2 (error diffusion)
8531 Frankie Sierra dithering v2 "Lite" (error diffusion)
8534 Default is @var{sierra2_4a}.
8537 When @var{bayer} dithering is selected, this option defines the scale of the
8538 pattern (how much the crosshatch pattern is visible). A low value means more
8539 visible pattern for less banding, and higher value means less visible pattern
8540 at the cost of more banding.
8542 The option must be an integer value in the range [0,5]. Default is @var{2}.
8545 If set, define the zone to process
8549 Only the changing rectangle will be reprocessed. This is similar to GIF
8550 cropping/offsetting compression mechanism. This option can be useful for speed
8551 if only a part of the image is changing, and has use cases such as limiting the
8552 scope of the error diffusal @option{dither} to the rectangle that bounds the
8553 moving scene (it leads to more deterministic output if the scene doesn't change
8554 much, and as a result less moving noise and better GIF compression).
8557 Default is @var{none}.
8560 @subsection Examples
8564 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
8565 using @command{ffmpeg}:
8567 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
8571 @section perspective
8573 Correct perspective of video not recorded perpendicular to the screen.
8575 A description of the accepted parameters follows.
8586 Set coordinates expression for top left, top right, bottom left and bottom right corners.
8587 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
8588 If the @code{sense} option is set to @code{source}, then the specified points will be sent
8589 to the corners of the destination. If the @code{sense} option is set to @code{destination},
8590 then the corners of the source will be sent to the specified coordinates.
8592 The expressions can use the following variables:
8597 the width and height of video frame.
8601 Set interpolation for perspective correction.
8603 It accepts the following values:
8609 Default value is @samp{linear}.
8612 Set interpretation of coordinate options.
8614 It accepts the following values:
8618 Send point in the source specified by the given coordinates to
8619 the corners of the destination.
8621 @item 1, destination
8623 Send the corners of the source to the point in the destination specified
8624 by the given coordinates.
8626 Default value is @samp{source}.
8632 Delay interlaced video by one field time so that the field order changes.
8634 The intended use is to fix PAL movies that have been captured with the
8635 opposite field order to the film-to-video transfer.
8637 A description of the accepted parameters follows.
8643 It accepts the following values:
8646 Capture field order top-first, transfer bottom-first.
8647 Filter will delay the bottom field.
8650 Capture field order bottom-first, transfer top-first.
8651 Filter will delay the top field.
8654 Capture and transfer with the same field order. This mode only exists
8655 for the documentation of the other options to refer to, but if you
8656 actually select it, the filter will faithfully do nothing.
8659 Capture field order determined automatically by field flags, transfer
8661 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
8662 basis using field flags. If no field information is available,
8663 then this works just like @samp{u}.
8666 Capture unknown or varying, transfer opposite.
8667 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
8668 analyzing the images and selecting the alternative that produces best
8669 match between the fields.
8672 Capture top-first, transfer unknown or varying.
8673 Filter selects among @samp{t} and @samp{p} using image analysis.
8676 Capture bottom-first, transfer unknown or varying.
8677 Filter selects among @samp{b} and @samp{p} using image analysis.
8680 Capture determined by field flags, transfer unknown or varying.
8681 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
8682 image analysis. If no field information is available, then this works just
8683 like @samp{U}. This is the default mode.
8686 Both capture and transfer unknown or varying.
8687 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
8691 @section pixdesctest
8693 Pixel format descriptor test filter, mainly useful for internal
8694 testing. The output video should be equal to the input video.
8698 format=monow, pixdesctest
8701 can be used to test the monowhite pixel format descriptor definition.
8705 Enable the specified chain of postprocessing subfilters using libpostproc. This
8706 library should be automatically selected with a GPL build (@code{--enable-gpl}).
8707 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
8708 Each subfilter and some options have a short and a long name that can be used
8709 interchangeably, i.e. dr/dering are the same.
8711 The filters accept the following options:
8715 Set postprocessing subfilters string.
8718 All subfilters share common options to determine their scope:
8722 Honor the quality commands for this subfilter.
8725 Do chrominance filtering, too (default).
8728 Do luminance filtering only (no chrominance).
8731 Do chrominance filtering only (no luminance).
8734 These options can be appended after the subfilter name, separated by a '|'.
8736 Available subfilters are:
8739 @item hb/hdeblock[|difference[|flatness]]
8740 Horizontal deblocking filter
8743 Difference factor where higher values mean more deblocking (default: @code{32}).
8745 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8748 @item vb/vdeblock[|difference[|flatness]]
8749 Vertical deblocking filter
8752 Difference factor where higher values mean more deblocking (default: @code{32}).
8754 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8757 @item ha/hadeblock[|difference[|flatness]]
8758 Accurate horizontal deblocking filter
8761 Difference factor where higher values mean more deblocking (default: @code{32}).
8763 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8766 @item va/vadeblock[|difference[|flatness]]
8767 Accurate vertical deblocking filter
8770 Difference factor where higher values mean more deblocking (default: @code{32}).
8772 Flatness threshold where lower values mean more deblocking (default: @code{39}).
8776 The horizontal and vertical deblocking filters share the difference and
8777 flatness values so you cannot set different horizontal and vertical
8782 Experimental horizontal deblocking filter
8785 Experimental vertical deblocking filter
8790 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
8793 larger -> stronger filtering
8795 larger -> stronger filtering
8797 larger -> stronger filtering
8800 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
8803 Stretch luminance to @code{0-255}.
8806 @item lb/linblenddeint
8807 Linear blend deinterlacing filter that deinterlaces the given block by
8808 filtering all lines with a @code{(1 2 1)} filter.
8810 @item li/linipoldeint
8811 Linear interpolating deinterlacing filter that deinterlaces the given block by
8812 linearly interpolating every second line.
8814 @item ci/cubicipoldeint
8815 Cubic interpolating deinterlacing filter deinterlaces the given block by
8816 cubically interpolating every second line.
8818 @item md/mediandeint
8819 Median deinterlacing filter that deinterlaces the given block by applying a
8820 median filter to every second line.
8822 @item fd/ffmpegdeint
8823 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
8824 second line with a @code{(-1 4 2 4 -1)} filter.
8827 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
8828 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
8830 @item fq/forceQuant[|quantizer]
8831 Overrides the quantizer table from the input with the constant quantizer you
8839 Default pp filter combination (@code{hb|a,vb|a,dr|a})
8842 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
8845 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
8848 @subsection Examples
8852 Apply horizontal and vertical deblocking, deringing and automatic
8853 brightness/contrast:
8859 Apply default filters without brightness/contrast correction:
8865 Apply default filters and temporal denoiser:
8867 pp=default/tmpnoise|1|2|3
8871 Apply deblocking on luminance only, and switch vertical deblocking on or off
8872 automatically depending on available CPU time:
8879 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
8880 similar to spp = 6 with 7 point DCT, where only the center sample is
8883 The filter accepts the following options:
8887 Force a constant quantization parameter. It accepts an integer in range
8888 0 to 63. If not set, the filter will use the QP from the video stream
8892 Set thresholding mode. Available modes are:
8896 Set hard thresholding.
8898 Set soft thresholding (better de-ringing effect, but likely blurrier).
8900 Set medium thresholding (good results, default).
8906 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
8907 Ratio) between two input videos.
8909 This filter takes in input two input videos, the first input is
8910 considered the "main" source and is passed unchanged to the
8911 output. The second input is used as a "reference" video for computing
8914 Both video inputs must have the same resolution and pixel format for
8915 this filter to work correctly. Also it assumes that both inputs
8916 have the same number of frames, which are compared one by one.
8918 The obtained average PSNR is printed through the logging system.
8920 The filter stores the accumulated MSE (mean squared error) of each
8921 frame, and at the end of the processing it is averaged across all frames
8922 equally, and the following formula is applied to obtain the PSNR:
8925 PSNR = 10*log10(MAX^2/MSE)
8928 Where MAX is the average of the maximum values of each component of the
8931 The description of the accepted parameters follows.
8935 If specified the filter will use the named file to save the PSNR of
8936 each individual frame.
8939 The file printed if @var{stats_file} is selected, contains a sequence of
8940 key/value pairs of the form @var{key}:@var{value} for each compared
8943 A description of each shown parameter follows:
8947 sequential number of the input frame, starting from 1
8950 Mean Square Error pixel-by-pixel average difference of the compared
8951 frames, averaged over all the image components.
8953 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
8954 Mean Square Error pixel-by-pixel average difference of the compared
8955 frames for the component specified by the suffix.
8957 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
8958 Peak Signal to Noise ratio of the compared frames for the component
8959 specified by the suffix.
8964 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
8965 [main][ref] psnr="stats_file=stats.log" [out]
8968 On this example the input file being processed is compared with the
8969 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
8970 is stored in @file{stats.log}.
8975 Pulldown reversal (inverse telecine) filter, capable of handling mixed
8976 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
8979 The pullup filter is designed to take advantage of future context in making
8980 its decisions. This filter is stateless in the sense that it does not lock
8981 onto a pattern to follow, but it instead looks forward to the following
8982 fields in order to identify matches and rebuild progressive frames.
8984 To produce content with an even framerate, insert the fps filter after
8985 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
8986 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
8988 The filter accepts the following options:
8995 These options set the amount of "junk" to ignore at the left, right, top, and
8996 bottom of the image, respectively. Left and right are in units of 8 pixels,
8997 while top and bottom are in units of 2 lines.
8998 The default is 8 pixels on each side.
9001 Set the strict breaks. Setting this option to 1 will reduce the chances of
9002 filter generating an occasional mismatched frame, but it may also cause an
9003 excessive number of frames to be dropped during high motion sequences.
9004 Conversely, setting it to -1 will make filter match fields more easily.
9005 This may help processing of video where there is slight blurring between
9006 the fields, but may also cause there to be interlaced frames in the output.
9007 Default value is @code{0}.
9010 Set the metric plane to use. It accepts the following values:
9016 Use chroma blue plane.
9019 Use chroma red plane.
9022 This option may be set to use chroma plane instead of the default luma plane
9023 for doing filter's computations. This may improve accuracy on very clean
9024 source material, but more likely will decrease accuracy, especially if there
9025 is chroma noise (rainbow effect) or any grayscale video.
9026 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
9027 load and make pullup usable in realtime on slow machines.
9030 For best results (without duplicated frames in the output file) it is
9031 necessary to change the output frame rate. For example, to inverse
9032 telecine NTSC input:
9034 ffmpeg -i input -vf pullup -r 24000/1001 ...
9039 Change video quantization parameters (QP).
9041 The filter accepts the following option:
9045 Set expression for quantization parameter.
9048 The expression is evaluated through the eval API and can contain, among others,
9049 the following constants:
9053 1 if index is not 129, 0 otherwise.
9056 Sequentional index starting from -129 to 128.
9059 @subsection Examples
9071 Flush video frames from internal cache of frames into a random order.
9072 No frame is discarded.
9073 Inspired by @ref{frei0r} nervous filter.
9077 Set size in number of frames of internal cache, in range from @code{2} to
9078 @code{512}. Default is @code{30}.
9081 Set seed for random number generator, must be an integer included between
9082 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
9083 less than @code{0}, the filter will try to use a good random seed on a
9087 @section removegrain
9089 The removegrain filter is a spatial denoiser for progressive video.
9093 Set mode for the first plane.
9096 Set mode for the second plane.
9099 Set mode for the third plane.
9102 Set mode for the fourth plane.
9105 Range of mode is from 0 to 24. Description of each mode follows:
9109 Leave input plane unchanged. Default.
9112 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
9115 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
9118 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
9121 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
9122 This is equivalent to a median filter.
9125 Line-sensitive clipping giving the minimal change.
9128 Line-sensitive clipping, intermediate.
9131 Line-sensitive clipping, intermediate.
9134 Line-sensitive clipping, intermediate.
9137 Line-sensitive clipping on a line where the neighbours pixels are the closest.
9140 Replaces the target pixel with the closest neighbour.
9143 [1 2 1] horizontal and vertical kernel blur.
9149 Bob mode, interpolates top field from the line where the neighbours
9150 pixels are the closest.
9153 Bob mode, interpolates bottom field from the line where the neighbours
9154 pixels are the closest.
9157 Bob mode, interpolates top field. Same as 13 but with a more complicated
9158 interpolation formula.
9161 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
9162 interpolation formula.
9165 Clips the pixel with the minimum and maximum of respectively the maximum and
9166 minimum of each pair of opposite neighbour pixels.
9169 Line-sensitive clipping using opposite neighbours whose greatest distance from
9170 the current pixel is minimal.
9173 Replaces the pixel with the average of its 8 neighbours.
9176 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
9179 Clips pixels using the averages of opposite neighbour.
9182 Same as mode 21 but simpler and faster.
9185 Small edge and halo removal, but reputed useless.
9193 Suppress a TV station logo, using an image file to determine which
9194 pixels comprise the logo. It works by filling in the pixels that
9195 comprise the logo with neighboring pixels.
9197 The filter accepts the following options:
9201 Set the filter bitmap file, which can be any image format supported by
9202 libavformat. The width and height of the image file must match those of the
9203 video stream being processed.
9206 Pixels in the provided bitmap image with a value of zero are not
9207 considered part of the logo, non-zero pixels are considered part of
9208 the logo. If you use white (255) for the logo and black (0) for the
9209 rest, you will be safe. For making the filter bitmap, it is
9210 recommended to take a screen capture of a black frame with the logo
9211 visible, and then using a threshold filter followed by the erode
9212 filter once or twice.
9214 If needed, little splotches can be fixed manually. Remember that if
9215 logo pixels are not covered, the filter quality will be much
9216 reduced. Marking too many pixels as part of the logo does not hurt as
9217 much, but it will increase the amount of blurring needed to cover over
9218 the image and will destroy more information than necessary, and extra
9219 pixels will slow things down on a large logo.
9221 @section repeatfields
9223 This filter uses the repeat_field flag from the Video ES headers and hard repeats
9224 fields based on its value.
9226 @section reverse, areverse
9230 Warning: This filter requires memory to buffer the entire clip, so trimming
9233 @subsection Examples
9237 Take the first 5 seconds of a clip, and reverse it.
9245 Rotate video by an arbitrary angle expressed in radians.
9247 The filter accepts the following options:
9249 A description of the optional parameters follows.
9252 Set an expression for the angle by which to rotate the input video
9253 clockwise, expressed as a number of radians. A negative value will
9254 result in a counter-clockwise rotation. By default it is set to "0".
9256 This expression is evaluated for each frame.
9259 Set the output width expression, default value is "iw".
9260 This expression is evaluated just once during configuration.
9263 Set the output height expression, default value is "ih".
9264 This expression is evaluated just once during configuration.
9267 Enable bilinear interpolation if set to 1, a value of 0 disables
9268 it. Default value is 1.
9271 Set the color used to fill the output area not covered by the rotated
9272 image. For the general syntax of this option, check the "Color" section in the
9273 ffmpeg-utils manual. If the special value "none" is selected then no
9274 background is printed (useful for example if the background is never shown).
9276 Default value is "black".
9279 The expressions for the angle and the output size can contain the
9280 following constants and functions:
9284 sequential number of the input frame, starting from 0. It is always NAN
9285 before the first frame is filtered.
9288 time in seconds of the input frame, it is set to 0 when the filter is
9289 configured. It is always NAN before the first frame is filtered.
9293 horizontal and vertical chroma subsample values. For example for the
9294 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9298 the input video width and height
9302 the output width and height, that is the size of the padded area as
9303 specified by the @var{width} and @var{height} expressions
9307 the minimal width/height required for completely containing the input
9308 video rotated by @var{a} radians.
9310 These are only available when computing the @option{out_w} and
9311 @option{out_h} expressions.
9314 @subsection Examples
9318 Rotate the input by PI/6 radians clockwise:
9324 Rotate the input by PI/6 radians counter-clockwise:
9330 Rotate the input by 45 degrees clockwise:
9336 Apply a constant rotation with period T, starting from an angle of PI/3:
9338 rotate=PI/3+2*PI*t/T
9342 Make the input video rotation oscillating with a period of T
9343 seconds and an amplitude of A radians:
9345 rotate=A*sin(2*PI/T*t)
9349 Rotate the video, output size is chosen so that the whole rotating
9350 input video is always completely contained in the output:
9352 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
9356 Rotate the video, reduce the output size so that no background is ever
9359 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
9363 @subsection Commands
9365 The filter supports the following commands:
9369 Set the angle expression.
9370 The command accepts the same syntax of the corresponding option.
9372 If the specified expression is not valid, it is kept at its current
9378 Apply Shape Adaptive Blur.
9380 The filter accepts the following options:
9383 @item luma_radius, lr
9384 Set luma blur filter strength, must be a value in range 0.1-4.0, default
9385 value is 1.0. A greater value will result in a more blurred image, and
9386 in slower processing.
9388 @item luma_pre_filter_radius, lpfr
9389 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
9392 @item luma_strength, ls
9393 Set luma maximum difference between pixels to still be considered, must
9394 be a value in the 0.1-100.0 range, default value is 1.0.
9396 @item chroma_radius, cr
9397 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
9398 greater value will result in a more blurred image, and in slower
9401 @item chroma_pre_filter_radius, cpfr
9402 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
9404 @item chroma_strength, cs
9405 Set chroma maximum difference between pixels to still be considered,
9406 must be a value in the 0.1-100.0 range.
9409 Each chroma option value, if not explicitly specified, is set to the
9410 corresponding luma option value.
9415 Scale (resize) the input video, using the libswscale library.
9417 The scale filter forces the output display aspect ratio to be the same
9418 of the input, by changing the output sample aspect ratio.
9420 If the input image format is different from the format requested by
9421 the next filter, the scale filter will convert the input to the
9425 The filter accepts the following options, or any of the options
9426 supported by the libswscale scaler.
9428 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
9429 the complete list of scaler options.
9434 Set the output video dimension expression. Default value is the input
9437 If the value is 0, the input width is used for the output.
9439 If one of the values is -1, the scale filter will use a value that
9440 maintains the aspect ratio of the input image, calculated from the
9441 other specified dimension. If both of them are -1, the input size is
9444 If one of the values is -n with n > 1, the scale filter will also use a value
9445 that maintains the aspect ratio of the input image, calculated from the other
9446 specified dimension. After that it will, however, make sure that the calculated
9447 dimension is divisible by n and adjust the value if necessary.
9449 See below for the list of accepted constants for use in the dimension
9453 Set the interlacing mode. It accepts the following values:
9457 Force interlaced aware scaling.
9460 Do not apply interlaced scaling.
9463 Select interlaced aware scaling depending on whether the source frames
9464 are flagged as interlaced or not.
9467 Default value is @samp{0}.
9470 Set libswscale scaling flags. See
9471 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
9472 complete list of values. If not explicitly specified the filter applies
9476 Set the video size. For the syntax of this option, check the
9477 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9479 @item in_color_matrix
9480 @item out_color_matrix
9481 Set in/output YCbCr color space type.
9483 This allows the autodetected value to be overridden as well as allows forcing
9484 a specific value used for the output and encoder.
9486 If not specified, the color space type depends on the pixel format.
9492 Choose automatically.
9495 Format conforming to International Telecommunication Union (ITU)
9496 Recommendation BT.709.
9499 Set color space conforming to the United States Federal Communications
9500 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
9503 Set color space conforming to:
9507 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
9510 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
9513 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
9518 Set color space conforming to SMPTE ST 240:1999.
9523 Set in/output YCbCr sample range.
9525 This allows the autodetected value to be overridden as well as allows forcing
9526 a specific value used for the output and encoder. If not specified, the
9527 range depends on the pixel format. Possible values:
9531 Choose automatically.
9534 Set full range (0-255 in case of 8-bit luma).
9537 Set "MPEG" range (16-235 in case of 8-bit luma).
9540 @item force_original_aspect_ratio
9541 Enable decreasing or increasing output video width or height if necessary to
9542 keep the original aspect ratio. Possible values:
9546 Scale the video as specified and disable this feature.
9549 The output video dimensions will automatically be decreased if needed.
9552 The output video dimensions will automatically be increased if needed.
9556 One useful instance of this option is that when you know a specific device's
9557 maximum allowed resolution, you can use this to limit the output video to
9558 that, while retaining the aspect ratio. For example, device A allows
9559 1280x720 playback, and your video is 1920x800. Using this option (set it to
9560 decrease) and specifying 1280x720 to the command line makes the output
9563 Please note that this is a different thing than specifying -1 for @option{w}
9564 or @option{h}, you still need to specify the output resolution for this option
9569 The values of the @option{w} and @option{h} options are expressions
9570 containing the following constants:
9575 The input width and height
9579 These are the same as @var{in_w} and @var{in_h}.
9583 The output (scaled) width and height
9587 These are the same as @var{out_w} and @var{out_h}
9590 The same as @var{iw} / @var{ih}
9593 input sample aspect ratio
9596 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
9600 horizontal and vertical input chroma subsample values. For example for the
9601 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9605 horizontal and vertical output chroma subsample values. For example for the
9606 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9609 @subsection Examples
9613 Scale the input video to a size of 200x100
9618 This is equivalent to:
9629 Specify a size abbreviation for the output size:
9634 which can also be written as:
9640 Scale the input to 2x:
9646 The above is the same as:
9652 Scale the input to 2x with forced interlaced scaling:
9654 scale=2*iw:2*ih:interl=1
9658 Scale the input to half size:
9664 Increase the width, and set the height to the same size:
9677 Increase the height, and set the width to 3/2 of the height:
9679 scale=w=3/2*oh:h=3/5*ih
9683 Increase the size, making the size a multiple of the chroma
9686 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
9690 Increase the width to a maximum of 500 pixels,
9691 keeping the same aspect ratio as the input:
9693 scale=w='min(500\, iw*3/2):h=-1'
9697 @subsection Commands
9699 This filter supports the following commands:
9703 Set the output video dimension expression.
9704 The command accepts the same syntax of the corresponding option.
9706 If the specified expression is not valid, it is kept at its current
9712 Scale (resize) the input video, based on a reference video.
9714 See the scale filter for available options, scale2ref supports the same but
9715 uses the reference video instead of the main input as basis.
9717 @subsection Examples
9721 Scale a subtitle stream to match the main video in size before overlaying
9723 'scale2ref[b][a];[a][b]overlay'
9727 @section separatefields
9729 The @code{separatefields} takes a frame-based video input and splits
9730 each frame into its components fields, producing a new half height clip
9731 with twice the frame rate and twice the frame count.
9733 This filter use field-dominance information in frame to decide which
9734 of each pair of fields to place first in the output.
9735 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
9737 @section setdar, setsar
9739 The @code{setdar} filter sets the Display Aspect Ratio for the filter
9742 This is done by changing the specified Sample (aka Pixel) Aspect
9743 Ratio, according to the following equation:
9745 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
9748 Keep in mind that the @code{setdar} filter does not modify the pixel
9749 dimensions of the video frame. Also, the display aspect ratio set by
9750 this filter may be changed by later filters in the filterchain,
9751 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
9754 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
9755 the filter output video.
9757 Note that as a consequence of the application of this filter, the
9758 output display aspect ratio will change according to the equation
9761 Keep in mind that the sample aspect ratio set by the @code{setsar}
9762 filter may be changed by later filters in the filterchain, e.g. if
9763 another "setsar" or a "setdar" filter is applied.
9765 It accepts the following parameters:
9768 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
9769 Set the aspect ratio used by the filter.
9771 The parameter can be a floating point number string, an expression, or
9772 a string of the form @var{num}:@var{den}, where @var{num} and
9773 @var{den} are the numerator and denominator of the aspect ratio. If
9774 the parameter is not specified, it is assumed the value "0".
9775 In case the form "@var{num}:@var{den}" is used, the @code{:} character
9779 Set the maximum integer value to use for expressing numerator and
9780 denominator when reducing the expressed aspect ratio to a rational.
9781 Default value is @code{100}.
9785 The parameter @var{sar} is an expression containing
9786 the following constants:
9790 These are approximated values for the mathematical constants e
9791 (Euler's number), pi (Greek pi), and phi (the golden ratio).
9794 The input width and height.
9797 These are the same as @var{w} / @var{h}.
9800 The input sample aspect ratio.
9803 The input display aspect ratio. It is the same as
9804 (@var{w} / @var{h}) * @var{sar}.
9807 Horizontal and vertical chroma subsample values. For example, for the
9808 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9811 @subsection Examples
9816 To change the display aspect ratio to 16:9, specify one of the following:
9824 To change the sample aspect ratio to 10:11, specify:
9830 To set a display aspect ratio of 16:9, and specify a maximum integer value of
9831 1000 in the aspect ratio reduction, use the command:
9833 setdar=ratio=16/9:max=1000
9841 Force field for the output video frame.
9843 The @code{setfield} filter marks the interlace type field for the
9844 output frames. It does not change the input frame, but only sets the
9845 corresponding property, which affects how the frame is treated by
9846 following filters (e.g. @code{fieldorder} or @code{yadif}).
9848 The filter accepts the following options:
9853 Available values are:
9857 Keep the same field property.
9860 Mark the frame as bottom-field-first.
9863 Mark the frame as top-field-first.
9866 Mark the frame as progressive.
9872 Show a line containing various information for each input video frame.
9873 The input video is not modified.
9875 The shown line contains a sequence of key/value pairs of the form
9876 @var{key}:@var{value}.
9878 The following values are shown in the output:
9882 The (sequential) number of the input frame, starting from 0.
9885 The Presentation TimeStamp of the input frame, expressed as a number of
9886 time base units. The time base unit depends on the filter input pad.
9889 The Presentation TimeStamp of the input frame, expressed as a number of
9893 The position of the frame in the input stream, or -1 if this information is
9894 unavailable and/or meaningless (for example in case of synthetic video).
9897 The pixel format name.
9900 The sample aspect ratio of the input frame, expressed in the form
9901 @var{num}/@var{den}.
9904 The size of the input frame. For the syntax of this option, check the
9905 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9908 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
9909 for bottom field first).
9912 This is 1 if the frame is a key frame, 0 otherwise.
9915 The picture type of the input frame ("I" for an I-frame, "P" for a
9916 P-frame, "B" for a B-frame, or "?" for an unknown type).
9917 Also refer to the documentation of the @code{AVPictureType} enum and of
9918 the @code{av_get_picture_type_char} function defined in
9919 @file{libavutil/avutil.h}.
9922 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
9924 @item plane_checksum
9925 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
9926 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
9929 @section showpalette
9931 Displays the 256 colors palette of each frame. This filter is only relevant for
9932 @var{pal8} pixel format frames.
9934 It accepts the following option:
9938 Set the size of the box used to represent one palette color entry. Default is
9939 @code{30} (for a @code{30x30} pixel box).
9942 @section shuffleplanes
9944 Reorder and/or duplicate video planes.
9946 It accepts the following parameters:
9951 The index of the input plane to be used as the first output plane.
9954 The index of the input plane to be used as the second output plane.
9957 The index of the input plane to be used as the third output plane.
9960 The index of the input plane to be used as the fourth output plane.
9964 The first plane has the index 0. The default is to keep the input unchanged.
9966 Swap the second and third planes of the input:
9968 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
9971 @anchor{signalstats}
9972 @section signalstats
9973 Evaluate various visual metrics that assist in determining issues associated
9974 with the digitization of analog video media.
9976 By default the filter will log these metadata values:
9980 Display the minimal Y value contained within the input frame. Expressed in
9984 Display the Y value at the 10% percentile within the input frame. Expressed in
9988 Display the average Y value within the input frame. Expressed in range of
9992 Display the Y value at the 90% percentile within the input frame. Expressed in
9996 Display the maximum Y value contained within the input frame. Expressed in
10000 Display the minimal U value contained within the input frame. Expressed in
10004 Display the U value at the 10% percentile within the input frame. Expressed in
10008 Display the average U value within the input frame. Expressed in range of
10012 Display the U value at the 90% percentile within the input frame. Expressed in
10016 Display the maximum U value contained within the input frame. Expressed in
10020 Display the minimal V value contained within the input frame. Expressed in
10024 Display the V value at the 10% percentile within the input frame. Expressed in
10028 Display the average V value within the input frame. Expressed in range of
10032 Display the V value at the 90% percentile within the input frame. Expressed in
10036 Display the maximum V value contained within the input frame. Expressed in
10040 Display the minimal saturation value contained within the input frame.
10041 Expressed in range of [0-~181.02].
10044 Display the saturation value at the 10% percentile within the input frame.
10045 Expressed in range of [0-~181.02].
10048 Display the average saturation value within the input frame. Expressed in range
10052 Display the saturation value at the 90% percentile within the input frame.
10053 Expressed in range of [0-~181.02].
10056 Display the maximum saturation value contained within the input frame.
10057 Expressed in range of [0-~181.02].
10060 Display the median value for hue within the input frame. Expressed in range of
10064 Display the average value for hue within the input frame. Expressed in range of
10068 Display the average of sample value difference between all values of the Y
10069 plane in the current frame and corresponding values of the previous input frame.
10070 Expressed in range of [0-255].
10073 Display the average of sample value difference between all values of the U
10074 plane in the current frame and corresponding values of the previous input frame.
10075 Expressed in range of [0-255].
10078 Display the average of sample value difference between all values of the V
10079 plane in the current frame and corresponding values of the previous input frame.
10080 Expressed in range of [0-255].
10083 The filter accepts the following options:
10089 @option{stat} specify an additional form of image analysis.
10090 @option{out} output video with the specified type of pixel highlighted.
10092 Both options accept the following values:
10096 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
10097 unlike the neighboring pixels of the same field. Examples of temporal outliers
10098 include the results of video dropouts, head clogs, or tape tracking issues.
10101 Identify @var{vertical line repetition}. Vertical line repetition includes
10102 similar rows of pixels within a frame. In born-digital video vertical line
10103 repetition is common, but this pattern is uncommon in video digitized from an
10104 analog source. When it occurs in video that results from the digitization of an
10105 analog source it can indicate concealment from a dropout compensator.
10108 Identify pixels that fall outside of legal broadcast range.
10112 Set the highlight color for the @option{out} option. The default color is
10116 @subsection Examples
10120 Output data of various video metrics:
10122 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
10126 Output specific data about the minimum and maximum values of the Y plane per frame:
10128 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
10132 Playback video while highlighting pixels that are outside of broadcast range in red.
10134 ffplay example.mov -vf signalstats="out=brng:color=red"
10138 Playback video with signalstats metadata drawn over the frame.
10140 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
10143 The contents of signalstat_drawtext.txt used in the command are:
10146 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
10147 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
10148 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
10149 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
10157 Blur the input video without impacting the outlines.
10159 It accepts the following options:
10162 @item luma_radius, lr
10163 Set the luma radius. The option value must be a float number in
10164 the range [0.1,5.0] that specifies the variance of the gaussian filter
10165 used to blur the image (slower if larger). Default value is 1.0.
10167 @item luma_strength, ls
10168 Set the luma strength. The option value must be a float number
10169 in the range [-1.0,1.0] that configures the blurring. A value included
10170 in [0.0,1.0] will blur the image whereas a value included in
10171 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10173 @item luma_threshold, lt
10174 Set the luma threshold used as a coefficient to determine
10175 whether a pixel should be blurred or not. The option value must be an
10176 integer in the range [-30,30]. A value of 0 will filter all the image,
10177 a value included in [0,30] will filter flat areas and a value included
10178 in [-30,0] will filter edges. Default value is 0.
10180 @item chroma_radius, cr
10181 Set the chroma radius. The option value must be a float number in
10182 the range [0.1,5.0] that specifies the variance of the gaussian filter
10183 used to blur the image (slower if larger). Default value is 1.0.
10185 @item chroma_strength, cs
10186 Set the chroma strength. The option value must be a float number
10187 in the range [-1.0,1.0] that configures the blurring. A value included
10188 in [0.0,1.0] will blur the image whereas a value included in
10189 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10191 @item chroma_threshold, ct
10192 Set the chroma threshold used as a coefficient to determine
10193 whether a pixel should be blurred or not. The option value must be an
10194 integer in the range [-30,30]. A value of 0 will filter all the image,
10195 a value included in [0,30] will filter flat areas and a value included
10196 in [-30,0] will filter edges. Default value is 0.
10199 If a chroma option is not explicitly set, the corresponding luma value
10204 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
10206 This filter takes in input two input videos, the first input is
10207 considered the "main" source and is passed unchanged to the
10208 output. The second input is used as a "reference" video for computing
10211 Both video inputs must have the same resolution and pixel format for
10212 this filter to work correctly. Also it assumes that both inputs
10213 have the same number of frames, which are compared one by one.
10215 The filter stores the calculated SSIM of each frame.
10217 The description of the accepted parameters follows.
10220 @item stats_file, f
10221 If specified the filter will use the named file to save the SSIM of
10222 each individual frame.
10225 The file printed if @var{stats_file} is selected, contains a sequence of
10226 key/value pairs of the form @var{key}:@var{value} for each compared
10229 A description of each shown parameter follows:
10233 sequential number of the input frame, starting from 1
10235 @item Y, U, V, R, G, B
10236 SSIM of the compared frames for the component specified by the suffix.
10239 SSIM of the compared frames for the whole frame.
10242 Same as above but in dB representation.
10247 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
10248 [main][ref] ssim="stats_file=stats.log" [out]
10251 On this example the input file being processed is compared with the
10252 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
10253 is stored in @file{stats.log}.
10255 Another example with both psnr and ssim at same time:
10257 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
10262 Convert between different stereoscopic image formats.
10264 The filters accept the following options:
10268 Set stereoscopic image format of input.
10270 Available values for input image formats are:
10273 side by side parallel (left eye left, right eye right)
10276 side by side crosseye (right eye left, left eye right)
10279 side by side parallel with half width resolution
10280 (left eye left, right eye right)
10283 side by side crosseye with half width resolution
10284 (right eye left, left eye right)
10287 above-below (left eye above, right eye below)
10290 above-below (right eye above, left eye below)
10293 above-below with half height resolution
10294 (left eye above, right eye below)
10297 above-below with half height resolution
10298 (right eye above, left eye below)
10301 alternating frames (left eye first, right eye second)
10304 alternating frames (right eye first, left eye second)
10307 interleaved rows (left eye has top row, right eye starts on next row)
10310 interleaved rows (right eye has top row, left eye starts on next row)
10312 Default value is @samp{sbsl}.
10316 Set stereoscopic image format of output.
10318 Available values for output image formats are all the input formats as well as:
10321 anaglyph red/blue gray
10322 (red filter on left eye, blue filter on right eye)
10325 anaglyph red/green gray
10326 (red filter on left eye, green filter on right eye)
10329 anaglyph red/cyan gray
10330 (red filter on left eye, cyan filter on right eye)
10333 anaglyph red/cyan half colored
10334 (red filter on left eye, cyan filter on right eye)
10337 anaglyph red/cyan color
10338 (red filter on left eye, cyan filter on right eye)
10341 anaglyph red/cyan color optimized with the least squares projection of dubois
10342 (red filter on left eye, cyan filter on right eye)
10345 anaglyph green/magenta gray
10346 (green filter on left eye, magenta filter on right eye)
10349 anaglyph green/magenta half colored
10350 (green filter on left eye, magenta filter on right eye)
10353 anaglyph green/magenta colored
10354 (green filter on left eye, magenta filter on right eye)
10357 anaglyph green/magenta color optimized with the least squares projection of dubois
10358 (green filter on left eye, magenta filter on right eye)
10361 anaglyph yellow/blue gray
10362 (yellow filter on left eye, blue filter on right eye)
10365 anaglyph yellow/blue half colored
10366 (yellow filter on left eye, blue filter on right eye)
10369 anaglyph yellow/blue colored
10370 (yellow filter on left eye, blue filter on right eye)
10373 anaglyph yellow/blue color optimized with the least squares projection of dubois
10374 (yellow filter on left eye, blue filter on right eye)
10377 mono output (left eye only)
10380 mono output (right eye only)
10383 checkerboard, left eye first
10386 checkerboard, right eye first
10389 interleaved columns, left eye first
10392 interleaved columns, right eye first
10395 Default value is @samp{arcd}.
10398 @subsection Examples
10402 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
10408 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
10417 Apply a simple postprocessing filter that compresses and decompresses the image
10418 at several (or - in the case of @option{quality} level @code{6} - all) shifts
10419 and average the results.
10421 The filter accepts the following options:
10425 Set quality. This option defines the number of levels for averaging. It accepts
10426 an integer in the range 0-6. If set to @code{0}, the filter will have no
10427 effect. A value of @code{6} means the higher quality. For each increment of
10428 that value the speed drops by a factor of approximately 2. Default value is
10432 Force a constant quantization parameter. If not set, the filter will use the QP
10433 from the video stream (if available).
10436 Set thresholding mode. Available modes are:
10440 Set hard thresholding (default).
10442 Set soft thresholding (better de-ringing effect, but likely blurrier).
10445 @item use_bframe_qp
10446 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
10447 option may cause flicker since the B-Frames have often larger QP. Default is
10448 @code{0} (not enabled).
10454 Draw subtitles on top of input video using the libass library.
10456 To enable compilation of this filter you need to configure FFmpeg with
10457 @code{--enable-libass}. This filter also requires a build with libavcodec and
10458 libavformat to convert the passed subtitles file to ASS (Advanced Substation
10459 Alpha) subtitles format.
10461 The filter accepts the following options:
10465 Set the filename of the subtitle file to read. It must be specified.
10467 @item original_size
10468 Specify the size of the original video, the video for which the ASS file
10469 was composed. For the syntax of this option, check the
10470 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10471 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
10472 correctly scale the fonts if the aspect ratio has been changed.
10475 Set a directory path containing fonts that can be used by the filter.
10476 These fonts will be used in addition to whatever the font provider uses.
10479 Set subtitles input character encoding. @code{subtitles} filter only. Only
10480 useful if not UTF-8.
10482 @item stream_index, si
10483 Set subtitles stream index. @code{subtitles} filter only.
10486 Override default style or script info parameters of the subtitles. It accepts a
10487 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
10490 If the first key is not specified, it is assumed that the first value
10491 specifies the @option{filename}.
10493 For example, to render the file @file{sub.srt} on top of the input
10494 video, use the command:
10499 which is equivalent to:
10501 subtitles=filename=sub.srt
10504 To render the default subtitles stream from file @file{video.mkv}, use:
10506 subtitles=video.mkv
10509 To render the second subtitles stream from that file, use:
10511 subtitles=video.mkv:si=1
10514 To make the subtitles stream from @file{sub.srt} appear in transparent green
10515 @code{DejaVu Serif}, use:
10517 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
10520 @section super2xsai
10522 Scale the input by 2x and smooth using the Super2xSaI (Scale and
10523 Interpolate) pixel art scaling algorithm.
10525 Useful for enlarging pixel art images without reducing sharpness.
10532 Apply telecine process to the video.
10534 This filter accepts the following options:
10543 The default value is @code{top}.
10547 A string of numbers representing the pulldown pattern you wish to apply.
10548 The default value is @code{23}.
10552 Some typical patterns:
10557 24p: 2332 (preferred)
10564 24p: 222222222223 ("Euro pulldown")
10570 Select the most representative frame in a given sequence of consecutive frames.
10572 The filter accepts the following options:
10576 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
10577 will pick one of them, and then handle the next batch of @var{n} frames until
10578 the end. Default is @code{100}.
10581 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
10582 value will result in a higher memory usage, so a high value is not recommended.
10584 @subsection Examples
10588 Extract one picture each 50 frames:
10594 Complete example of a thumbnail creation with @command{ffmpeg}:
10596 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
10602 Tile several successive frames together.
10604 The filter accepts the following options:
10609 Set the grid size (i.e. the number of lines and columns). For the syntax of
10610 this option, check the
10611 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10614 Set the maximum number of frames to render in the given area. It must be less
10615 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
10616 the area will be used.
10619 Set the outer border margin in pixels.
10622 Set the inner border thickness (i.e. the number of pixels between frames). For
10623 more advanced padding options (such as having different values for the edges),
10624 refer to the pad video filter.
10627 Specify the color of the unused area. For the syntax of this option, check the
10628 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
10632 @subsection Examples
10636 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
10638 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
10640 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
10641 duplicating each output frame to accommodate the originally detected frame
10645 Display @code{5} pictures in an area of @code{3x2} frames,
10646 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
10647 mixed flat and named options:
10649 tile=3x2:nb_frames=5:padding=7:margin=2
10653 @section tinterlace
10655 Perform various types of temporal field interlacing.
10657 Frames are counted starting from 1, so the first input frame is
10660 The filter accepts the following options:
10665 Specify the mode of the interlacing. This option can also be specified
10666 as a value alone. See below for a list of values for this option.
10668 Available values are:
10672 Move odd frames into the upper field, even into the lower field,
10673 generating a double height frame at half frame rate.
10677 Frame 1 Frame 2 Frame 3 Frame 4
10679 11111 22222 33333 44444
10680 11111 22222 33333 44444
10681 11111 22222 33333 44444
10682 11111 22222 33333 44444
10696 Only output even frames, odd frames are dropped, generating a frame with
10697 unchanged height at half frame rate.
10702 Frame 1 Frame 2 Frame 3 Frame 4
10704 11111 22222 33333 44444
10705 11111 22222 33333 44444
10706 11111 22222 33333 44444
10707 11111 22222 33333 44444
10717 Only output odd frames, even frames are dropped, generating a frame with
10718 unchanged height at half frame rate.
10723 Frame 1 Frame 2 Frame 3 Frame 4
10725 11111 22222 33333 44444
10726 11111 22222 33333 44444
10727 11111 22222 33333 44444
10728 11111 22222 33333 44444
10738 Expand each frame to full height, but pad alternate lines with black,
10739 generating a frame with double height at the same input frame rate.
10744 Frame 1 Frame 2 Frame 3 Frame 4
10746 11111 22222 33333 44444
10747 11111 22222 33333 44444
10748 11111 22222 33333 44444
10749 11111 22222 33333 44444
10752 11111 ..... 33333 .....
10753 ..... 22222 ..... 44444
10754 11111 ..... 33333 .....
10755 ..... 22222 ..... 44444
10756 11111 ..... 33333 .....
10757 ..... 22222 ..... 44444
10758 11111 ..... 33333 .....
10759 ..... 22222 ..... 44444
10763 @item interleave_top, 4
10764 Interleave the upper field from odd frames with the lower field from
10765 even frames, generating a frame with unchanged height at half frame rate.
10770 Frame 1 Frame 2 Frame 3 Frame 4
10772 11111<- 22222 33333<- 44444
10773 11111 22222<- 33333 44444<-
10774 11111<- 22222 33333<- 44444
10775 11111 22222<- 33333 44444<-
10785 @item interleave_bottom, 5
10786 Interleave the lower field from odd frames with the upper field from
10787 even frames, generating a frame with unchanged height at half frame rate.
10792 Frame 1 Frame 2 Frame 3 Frame 4
10794 11111 22222<- 33333 44444<-
10795 11111<- 22222 33333<- 44444
10796 11111 22222<- 33333 44444<-
10797 11111<- 22222 33333<- 44444
10807 @item interlacex2, 6
10808 Double frame rate with unchanged height. Frames are inserted each
10809 containing the second temporal field from the previous input frame and
10810 the first temporal field from the next input frame. This mode relies on
10811 the top_field_first flag. Useful for interlaced video displays with no
10812 field synchronisation.
10817 Frame 1 Frame 2 Frame 3 Frame 4
10819 11111 22222 33333 44444
10820 11111 22222 33333 44444
10821 11111 22222 33333 44444
10822 11111 22222 33333 44444
10825 11111 22222 22222 33333 33333 44444 44444
10826 11111 11111 22222 22222 33333 33333 44444
10827 11111 22222 22222 33333 33333 44444 44444
10828 11111 11111 22222 22222 33333 33333 44444
10832 Move odd frames into the upper field, even into the lower field,
10833 generating a double height frame at same frame rate.
10837 Frame 1 Frame 2 Frame 3 Frame 4
10839 11111 22222 33333 44444
10840 11111 22222 33333 44444
10841 11111 22222 33333 44444
10842 11111 22222 33333 44444
10845 11111 33333 33333 55555
10846 22222 22222 44444 44444
10847 11111 33333 33333 55555
10848 22222 22222 44444 44444
10849 11111 33333 33333 55555
10850 22222 22222 44444 44444
10851 11111 33333 33333 55555
10852 22222 22222 44444 44444
10857 Numeric values are deprecated but are accepted for backward
10858 compatibility reasons.
10860 Default mode is @code{merge}.
10863 Specify flags influencing the filter process.
10865 Available value for @var{flags} is:
10868 @item low_pass_filter, vlfp
10869 Enable vertical low-pass filtering in the filter.
10870 Vertical low-pass filtering is required when creating an interlaced
10871 destination from a progressive source which contains high-frequency
10872 vertical detail. Filtering will reduce interlace 'twitter' and Moire
10875 Vertical low-pass filtering can only be enabled for @option{mode}
10876 @var{interleave_top} and @var{interleave_bottom}.
10883 Transpose rows with columns in the input video and optionally flip it.
10885 It accepts the following parameters:
10890 Specify the transposition direction.
10892 Can assume the following values:
10894 @item 0, 4, cclock_flip
10895 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
10903 Rotate by 90 degrees clockwise, that is:
10911 Rotate by 90 degrees counterclockwise, that is:
10918 @item 3, 7, clock_flip
10919 Rotate by 90 degrees clockwise and vertically flip, that is:
10927 For values between 4-7, the transposition is only done if the input
10928 video geometry is portrait and not landscape. These values are
10929 deprecated, the @code{passthrough} option should be used instead.
10931 Numerical values are deprecated, and should be dropped in favor of
10932 symbolic constants.
10935 Do not apply the transposition if the input geometry matches the one
10936 specified by the specified value. It accepts the following values:
10939 Always apply transposition.
10941 Preserve portrait geometry (when @var{height} >= @var{width}).
10943 Preserve landscape geometry (when @var{width} >= @var{height}).
10946 Default value is @code{none}.
10949 For example to rotate by 90 degrees clockwise and preserve portrait
10952 transpose=dir=1:passthrough=portrait
10955 The command above can also be specified as:
10957 transpose=1:portrait
10961 Trim the input so that the output contains one continuous subpart of the input.
10963 It accepts the following parameters:
10966 Specify the time of the start of the kept section, i.e. the frame with the
10967 timestamp @var{start} will be the first frame in the output.
10970 Specify the time of the first frame that will be dropped, i.e. the frame
10971 immediately preceding the one with the timestamp @var{end} will be the last
10972 frame in the output.
10975 This is the same as @var{start}, except this option sets the start timestamp
10976 in timebase units instead of seconds.
10979 This is the same as @var{end}, except this option sets the end timestamp
10980 in timebase units instead of seconds.
10983 The maximum duration of the output in seconds.
10986 The number of the first frame that should be passed to the output.
10989 The number of the first frame that should be dropped.
10992 @option{start}, @option{end}, and @option{duration} are expressed as time
10993 duration specifications; see
10994 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
10995 for the accepted syntax.
10997 Note that the first two sets of the start/end options and the @option{duration}
10998 option look at the frame timestamp, while the _frame variants simply count the
10999 frames that pass through the filter. Also note that this filter does not modify
11000 the timestamps. If you wish for the output timestamps to start at zero, insert a
11001 setpts filter after the trim filter.
11003 If multiple start or end options are set, this filter tries to be greedy and
11004 keep all the frames that match at least one of the specified constraints. To keep
11005 only the part that matches all the constraints at once, chain multiple trim
11008 The defaults are such that all the input is kept. So it is possible to set e.g.
11009 just the end values to keep everything before the specified time.
11014 Drop everything except the second minute of input:
11016 ffmpeg -i INPUT -vf trim=60:120
11020 Keep only the first second:
11022 ffmpeg -i INPUT -vf trim=duration=1
11031 Sharpen or blur the input video.
11033 It accepts the following parameters:
11036 @item luma_msize_x, lx
11037 Set the luma matrix horizontal size. It must be an odd integer between
11038 3 and 63. The default value is 5.
11040 @item luma_msize_y, ly
11041 Set the luma matrix vertical size. It must be an odd integer between 3
11042 and 63. The default value is 5.
11044 @item luma_amount, la
11045 Set the luma effect strength. It must be a floating point number, reasonable
11046 values lay between -1.5 and 1.5.
11048 Negative values will blur the input video, while positive values will
11049 sharpen it, a value of zero will disable the effect.
11051 Default value is 1.0.
11053 @item chroma_msize_x, cx
11054 Set the chroma matrix horizontal size. It must be an odd integer
11055 between 3 and 63. The default value is 5.
11057 @item chroma_msize_y, cy
11058 Set the chroma matrix vertical size. It must be an odd integer
11059 between 3 and 63. The default value is 5.
11061 @item chroma_amount, ca
11062 Set the chroma effect strength. It must be a floating point number, reasonable
11063 values lay between -1.5 and 1.5.
11065 Negative values will blur the input video, while positive values will
11066 sharpen it, a value of zero will disable the effect.
11068 Default value is 0.0.
11071 If set to 1, specify using OpenCL capabilities, only available if
11072 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
11076 All parameters are optional and default to the equivalent of the
11077 string '5:5:1.0:5:5:0.0'.
11079 @subsection Examples
11083 Apply strong luma sharpen effect:
11085 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
11089 Apply a strong blur of both luma and chroma parameters:
11091 unsharp=7:7:-2:7:7:-2
11097 Apply ultra slow/simple postprocessing filter that compresses and decompresses
11098 the image at several (or - in the case of @option{quality} level @code{8} - all)
11099 shifts and average the results.
11101 The way this differs from the behavior of spp is that uspp actually encodes &
11102 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
11103 DCT similar to MJPEG.
11105 The filter accepts the following options:
11109 Set quality. This option defines the number of levels for averaging. It accepts
11110 an integer in the range 0-8. If set to @code{0}, the filter will have no
11111 effect. A value of @code{8} means the higher quality. For each increment of
11112 that value the speed drops by a factor of approximately 2. Default value is
11116 Force a constant quantization parameter. If not set, the filter will use the QP
11117 from the video stream (if available).
11120 @section vectorscope
11122 Display 2 color component values in the two dimensional graph (which is called
11125 This filter accepts the following options:
11129 Set vectorscope mode.
11131 It accepts the following values:
11134 Gray values are displayed on graph, higher brightness means more pixels have
11135 same component color value on location in graph. This is the default mode.
11138 Gray values are displayed on graph. Surrounding pixels values which are not
11139 present in video frame are drawn in gradient of 2 color components which are
11140 set by option @code{x} and @code{y}.
11143 Actual color components values present in video frame are displayed on graph.
11146 Similar as color2 but higher frequency of same values @code{x} and @code{y}
11147 on graph increases value of another color component, which is luminance by
11148 default values of @code{x} and @code{y}.
11151 Actual colors present in video frame are displayed on graph. If two different
11152 colors map to same position on graph then color with higher value of component
11153 not present in graph is picked.
11157 Set which color component will be represented on X-axis. Default is @code{1}.
11160 Set which color component will be represented on Y-axis. Default is @code{2}.
11163 Set intensity, used by modes: gray, color and color3 for increasing brightness
11164 of color component which represents frequency of (X, Y) location in graph.
11169 No envelope, this is default.
11172 Instant envelope, even darkest single pixel will be clearly highlighted.
11175 Hold maximum and minimum values presented in graph over time. This way you
11176 can still spot out of range values without constantly looking at vectorscope.
11179 Peak and instant envelope combined together.
11183 @anchor{vidstabdetect}
11184 @section vidstabdetect
11186 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
11187 @ref{vidstabtransform} for pass 2.
11189 This filter generates a file with relative translation and rotation
11190 transform information about subsequent frames, which is then used by
11191 the @ref{vidstabtransform} filter.
11193 To enable compilation of this filter you need to configure FFmpeg with
11194 @code{--enable-libvidstab}.
11196 This filter accepts the following options:
11200 Set the path to the file used to write the transforms information.
11201 Default value is @file{transforms.trf}.
11204 Set how shaky the video is and how quick the camera is. It accepts an
11205 integer in the range 1-10, a value of 1 means little shakiness, a
11206 value of 10 means strong shakiness. Default value is 5.
11209 Set the accuracy of the detection process. It must be a value in the
11210 range 1-15. A value of 1 means low accuracy, a value of 15 means high
11211 accuracy. Default value is 15.
11214 Set stepsize of the search process. The region around minimum is
11215 scanned with 1 pixel resolution. Default value is 6.
11218 Set minimum contrast. Below this value a local measurement field is
11219 discarded. Must be a floating point value in the range 0-1. Default
11223 Set reference frame number for tripod mode.
11225 If enabled, the motion of the frames is compared to a reference frame
11226 in the filtered stream, identified by the specified number. The idea
11227 is to compensate all movements in a more-or-less static scene and keep
11228 the camera view absolutely still.
11230 If set to 0, it is disabled. The frames are counted starting from 1.
11233 Show fields and transforms in the resulting frames. It accepts an
11234 integer in the range 0-2. Default value is 0, which disables any
11238 @subsection Examples
11242 Use default values:
11248 Analyze strongly shaky movie and put the results in file
11249 @file{mytransforms.trf}:
11251 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
11255 Visualize the result of internal transformations in the resulting
11258 vidstabdetect=show=1
11262 Analyze a video with medium shakiness using @command{ffmpeg}:
11264 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
11268 @anchor{vidstabtransform}
11269 @section vidstabtransform
11271 Video stabilization/deshaking: pass 2 of 2,
11272 see @ref{vidstabdetect} for pass 1.
11274 Read a file with transform information for each frame and
11275 apply/compensate them. Together with the @ref{vidstabdetect}
11276 filter this can be used to deshake videos. See also
11277 @url{http://public.hronopik.de/vid.stab}. It is important to also use
11278 the @ref{unsharp} filter, see below.
11280 To enable compilation of this filter you need to configure FFmpeg with
11281 @code{--enable-libvidstab}.
11283 @subsection Options
11287 Set path to the file used to read the transforms. Default value is
11288 @file{transforms.trf}.
11291 Set the number of frames (value*2 + 1) used for lowpass filtering the
11292 camera movements. Default value is 10.
11294 For example a number of 10 means that 21 frames are used (10 in the
11295 past and 10 in the future) to smoothen the motion in the video. A
11296 larger value leads to a smoother video, but limits the acceleration of
11297 the camera (pan/tilt movements). 0 is a special case where a static
11298 camera is simulated.
11301 Set the camera path optimization algorithm.
11303 Accepted values are:
11306 gaussian kernel low-pass filter on camera motion (default)
11308 averaging on transformations
11312 Set maximal number of pixels to translate frames. Default value is -1,
11316 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
11317 value is -1, meaning no limit.
11320 Specify how to deal with borders that may be visible due to movement
11323 Available values are:
11326 keep image information from previous frame (default)
11328 fill the border black
11332 Invert transforms if set to 1. Default value is 0.
11335 Consider transforms as relative to previous frame if set to 1,
11336 absolute if set to 0. Default value is 0.
11339 Set percentage to zoom. A positive value will result in a zoom-in
11340 effect, a negative value in a zoom-out effect. Default value is 0 (no
11344 Set optimal zooming to avoid borders.
11346 Accepted values are:
11351 optimal static zoom value is determined (only very strong movements
11352 will lead to visible borders) (default)
11354 optimal adaptive zoom value is determined (no borders will be
11355 visible), see @option{zoomspeed}
11358 Note that the value given at zoom is added to the one calculated here.
11361 Set percent to zoom maximally each frame (enabled when
11362 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
11366 Specify type of interpolation.
11368 Available values are:
11373 linear only horizontal
11375 linear in both directions (default)
11377 cubic in both directions (slow)
11381 Enable virtual tripod mode if set to 1, which is equivalent to
11382 @code{relative=0:smoothing=0}. Default value is 0.
11384 Use also @code{tripod} option of @ref{vidstabdetect}.
11387 Increase log verbosity if set to 1. Also the detected global motions
11388 are written to the temporary file @file{global_motions.trf}. Default
11392 @subsection Examples
11396 Use @command{ffmpeg} for a typical stabilization with default values:
11398 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
11401 Note the use of the @ref{unsharp} filter which is always recommended.
11404 Zoom in a bit more and load transform data from a given file:
11406 vidstabtransform=zoom=5:input="mytransforms.trf"
11410 Smoothen the video even more:
11412 vidstabtransform=smoothing=30
11418 Flip the input video vertically.
11420 For example, to vertically flip a video with @command{ffmpeg}:
11422 ffmpeg -i in.avi -vf "vflip" out.avi
11428 Make or reverse a natural vignetting effect.
11430 The filter accepts the following options:
11434 Set lens angle expression as a number of radians.
11436 The value is clipped in the @code{[0,PI/2]} range.
11438 Default value: @code{"PI/5"}
11442 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
11446 Set forward/backward mode.
11448 Available modes are:
11451 The larger the distance from the central point, the darker the image becomes.
11454 The larger the distance from the central point, the brighter the image becomes.
11455 This can be used to reverse a vignette effect, though there is no automatic
11456 detection to extract the lens @option{angle} and other settings (yet). It can
11457 also be used to create a burning effect.
11460 Default value is @samp{forward}.
11463 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
11465 It accepts the following values:
11468 Evaluate expressions only once during the filter initialization.
11471 Evaluate expressions for each incoming frame. This is way slower than the
11472 @samp{init} mode since it requires all the scalers to be re-computed, but it
11473 allows advanced dynamic expressions.
11476 Default value is @samp{init}.
11479 Set dithering to reduce the circular banding effects. Default is @code{1}
11483 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
11484 Setting this value to the SAR of the input will make a rectangular vignetting
11485 following the dimensions of the video.
11487 Default is @code{1/1}.
11490 @subsection Expressions
11492 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
11493 following parameters.
11498 input width and height
11501 the number of input frame, starting from 0
11504 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
11505 @var{TB} units, NAN if undefined
11508 frame rate of the input video, NAN if the input frame rate is unknown
11511 the PTS (Presentation TimeStamp) of the filtered video frame,
11512 expressed in seconds, NAN if undefined
11515 time base of the input video
11519 @subsection Examples
11523 Apply simple strong vignetting effect:
11529 Make a flickering vignetting:
11531 vignette='PI/4+random(1)*PI/50':eval=frame
11537 Stack input videos vertically.
11539 All streams must be of same pixel format and of same width.
11541 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
11542 to create same output.
11544 The filter accept the following option:
11548 Set number of input streams. Default is 2.
11553 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
11554 Deinterlacing Filter").
11556 Based on the process described by Martin Weston for BBC R&D, and
11557 implemented based on the de-interlace algorithm written by Jim
11558 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
11559 uses filter coefficients calculated by BBC R&D.
11561 There are two sets of filter coefficients, so called "simple":
11562 and "complex". Which set of filter coefficients is used can
11563 be set by passing an optional parameter:
11567 Set the interlacing filter coefficients. Accepts one of the following values:
11571 Simple filter coefficient set.
11573 More-complex filter coefficient set.
11575 Default value is @samp{complex}.
11578 Specify which frames to deinterlace. Accept one of the following values:
11582 Deinterlace all frames,
11584 Only deinterlace frames marked as interlaced.
11587 Default value is @samp{all}.
11591 Video waveform monitor.
11593 The waveform monitor plots color component intensity. By default luminance
11594 only. Each column of the waveform corresponds to a column of pixels in the
11597 It accepts the following options:
11601 Can be either @code{row}, or @code{column}. Default is @code{column}.
11602 In row mode, the graph on the left side represents color component value 0 and
11603 the right side represents value = 255. In column mode, the top side represents
11604 color component value = 0 and bottom side represents value = 255.
11607 Set intensity. Smaller values are useful to find out how many values of the same
11608 luminance are distributed across input rows/columns.
11609 Default value is @code{0.04}. Allowed range is [0, 1].
11612 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
11613 In mirrored mode, higher values will be represented on the left
11614 side for @code{row} mode and at the top for @code{column} mode. Default is
11615 @code{1} (mirrored).
11619 It accepts the following values:
11622 Presents information identical to that in the @code{parade}, except
11623 that the graphs representing color components are superimposed directly
11626 This display mode makes it easier to spot relative differences or similarities
11627 in overlapping areas of the color components that are supposed to be identical,
11628 such as neutral whites, grays, or blacks.
11631 Display separate graph for the color components side by side in
11632 @code{row} mode or one below the other in @code{column} mode.
11634 Using this display mode makes it easy to spot color casts in the highlights
11635 and shadows of an image, by comparing the contours of the top and the bottom
11636 graphs of each waveform. Since whites, grays, and blacks are characterized
11637 by exactly equal amounts of red, green, and blue, neutral areas of the picture
11638 should display three waveforms of roughly equal width/height. If not, the
11639 correction is easy to perform by making level adjustments the three waveforms.
11641 Default is @code{parade}.
11643 @item components, c
11644 Set which color components to display. Default is 1, which means only luminance
11645 or red color component if input is in RGB colorspace. If is set for example to
11646 7 it will display all 3 (if) available color components.
11651 No envelope, this is default.
11654 Instant envelope, minimum and maximum values presented in graph will be easily
11655 visible even with small @code{step} value.
11658 Hold minimum and maximum values presented in graph across time. This way you
11659 can still spot out of range values without constantly looking at waveforms.
11662 Peak and instant envelope combined together.
11668 No filtering, this is default.
11671 Luma and chroma combined together.
11674 Similar as above, but shows difference between blue and red chroma.
11677 Displays only chroma.
11680 Similar as above, but shows difference between blue and red chroma.
11683 Displays actual color value on waveform.
11688 Apply the xBR high-quality magnification filter which is designed for pixel
11689 art. It follows a set of edge-detection rules, see
11690 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
11692 It accepts the following option:
11696 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
11697 @code{3xBR} and @code{4} for @code{4xBR}.
11698 Default is @code{3}.
11704 Deinterlace the input video ("yadif" means "yet another deinterlacing
11707 It accepts the following parameters:
11713 The interlacing mode to adopt. It accepts one of the following values:
11716 @item 0, send_frame
11717 Output one frame for each frame.
11718 @item 1, send_field
11719 Output one frame for each field.
11720 @item 2, send_frame_nospatial
11721 Like @code{send_frame}, but it skips the spatial interlacing check.
11722 @item 3, send_field_nospatial
11723 Like @code{send_field}, but it skips the spatial interlacing check.
11726 The default value is @code{send_frame}.
11729 The picture field parity assumed for the input interlaced video. It accepts one
11730 of the following values:
11734 Assume the top field is first.
11736 Assume the bottom field is first.
11738 Enable automatic detection of field parity.
11741 The default value is @code{auto}.
11742 If the interlacing is unknown or the decoder does not export this information,
11743 top field first will be assumed.
11746 Specify which frames to deinterlace. Accept one of the following
11751 Deinterlace all frames.
11752 @item 1, interlaced
11753 Only deinterlace frames marked as interlaced.
11756 The default value is @code{all}.
11761 Apply Zoom & Pan effect.
11763 This filter accepts the following options:
11767 Set the zoom expression. Default is 1.
11771 Set the x and y expression. Default is 0.
11774 Set the duration expression in number of frames.
11775 This sets for how many number of frames effect will last for
11776 single input image.
11779 Set the output image size, default is 'hd720'.
11782 Each expression can contain the following constants:
11801 Output frame count.
11805 Last calculated 'x' and 'y' position from 'x' and 'y' expression
11806 for current input frame.
11810 'x' and 'y' of last output frame of previous input frame or 0 when there was
11811 not yet such frame (first input frame).
11814 Last calculated zoom from 'z' expression for current input frame.
11817 Last calculated zoom of last output frame of previous input frame.
11820 Number of output frames for current input frame. Calculated from 'd' expression
11821 for each input frame.
11824 number of output frames created for previous input frame
11827 Rational number: input width / input height
11830 sample aspect ratio
11833 display aspect ratio
11837 @subsection Examples
11841 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
11843 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
11847 Zoom-in up to 1.5 and pan always at center of picture:
11849 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
11853 @c man end VIDEO FILTERS
11855 @chapter Video Sources
11856 @c man begin VIDEO SOURCES
11858 Below is a description of the currently available video sources.
11862 Buffer video frames, and make them available to the filter chain.
11864 This source is mainly intended for a programmatic use, in particular
11865 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
11867 It accepts the following parameters:
11872 Specify the size (width and height) of the buffered video frames. For the
11873 syntax of this option, check the
11874 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11877 The input video width.
11880 The input video height.
11883 A string representing the pixel format of the buffered video frames.
11884 It may be a number corresponding to a pixel format, or a pixel format
11888 Specify the timebase assumed by the timestamps of the buffered frames.
11891 Specify the frame rate expected for the video stream.
11893 @item pixel_aspect, sar
11894 The sample (pixel) aspect ratio of the input video.
11897 Specify the optional parameters to be used for the scale filter which
11898 is automatically inserted when an input change is detected in the
11899 input size or format.
11904 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
11907 will instruct the source to accept video frames with size 320x240 and
11908 with format "yuv410p", assuming 1/24 as the timestamps timebase and
11909 square pixels (1:1 sample aspect ratio).
11910 Since the pixel format with name "yuv410p" corresponds to the number 6
11911 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
11912 this example corresponds to:
11914 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
11917 Alternatively, the options can be specified as a flat string, but this
11918 syntax is deprecated:
11920 @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}]
11924 Create a pattern generated by an elementary cellular automaton.
11926 The initial state of the cellular automaton can be defined through the
11927 @option{filename}, and @option{pattern} options. If such options are
11928 not specified an initial state is created randomly.
11930 At each new frame a new row in the video is filled with the result of
11931 the cellular automaton next generation. The behavior when the whole
11932 frame is filled is defined by the @option{scroll} option.
11934 This source accepts the following options:
11938 Read the initial cellular automaton state, i.e. the starting row, from
11939 the specified file.
11940 In the file, each non-whitespace character is considered an alive
11941 cell, a newline will terminate the row, and further characters in the
11942 file will be ignored.
11945 Read the initial cellular automaton state, i.e. the starting row, from
11946 the specified string.
11948 Each non-whitespace character in the string is considered an alive
11949 cell, a newline will terminate the row, and further characters in the
11950 string will be ignored.
11953 Set the video rate, that is the number of frames generated per second.
11956 @item random_fill_ratio, ratio
11957 Set the random fill ratio for the initial cellular automaton row. It
11958 is a floating point number value ranging from 0 to 1, defaults to
11961 This option is ignored when a file or a pattern is specified.
11963 @item random_seed, seed
11964 Set the seed for filling randomly the initial row, must be an integer
11965 included between 0 and UINT32_MAX. If not specified, or if explicitly
11966 set to -1, the filter will try to use a good random seed on a best
11970 Set the cellular automaton rule, it is a number ranging from 0 to 255.
11971 Default value is 110.
11974 Set the size of the output video. For the syntax of this option, check the
11975 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11977 If @option{filename} or @option{pattern} is specified, the size is set
11978 by default to the width of the specified initial state row, and the
11979 height is set to @var{width} * PHI.
11981 If @option{size} is set, it must contain the width of the specified
11982 pattern string, and the specified pattern will be centered in the
11985 If a filename or a pattern string is not specified, the size value
11986 defaults to "320x518" (used for a randomly generated initial state).
11989 If set to 1, scroll the output upward when all the rows in the output
11990 have been already filled. If set to 0, the new generated row will be
11991 written over the top row just after the bottom row is filled.
11994 @item start_full, full
11995 If set to 1, completely fill the output with generated rows before
11996 outputting the first frame.
11997 This is the default behavior, for disabling set the value to 0.
12000 If set to 1, stitch the left and right row edges together.
12001 This is the default behavior, for disabling set the value to 0.
12004 @subsection Examples
12008 Read the initial state from @file{pattern}, and specify an output of
12011 cellauto=f=pattern:s=200x400
12015 Generate a random initial row with a width of 200 cells, with a fill
12018 cellauto=ratio=2/3:s=200x200
12022 Create a pattern generated by rule 18 starting by a single alive cell
12023 centered on an initial row with width 100:
12025 cellauto=p=@@:s=100x400:full=0:rule=18
12029 Specify a more elaborated initial pattern:
12031 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
12036 @section mandelbrot
12038 Generate a Mandelbrot set fractal, and progressively zoom towards the
12039 point specified with @var{start_x} and @var{start_y}.
12041 This source accepts the following options:
12046 Set the terminal pts value. Default value is 400.
12049 Set the terminal scale value.
12050 Must be a floating point value. Default value is 0.3.
12053 Set the inner coloring mode, that is the algorithm used to draw the
12054 Mandelbrot fractal internal region.
12056 It shall assume one of the following values:
12061 Show time until convergence.
12063 Set color based on point closest to the origin of the iterations.
12068 Default value is @var{mincol}.
12071 Set the bailout value. Default value is 10.0.
12074 Set the maximum of iterations performed by the rendering
12075 algorithm. Default value is 7189.
12078 Set outer coloring mode.
12079 It shall assume one of following values:
12081 @item iteration_count
12082 Set iteration cound mode.
12083 @item normalized_iteration_count
12084 set normalized iteration count mode.
12086 Default value is @var{normalized_iteration_count}.
12089 Set frame rate, expressed as number of frames per second. Default
12093 Set frame size. For the syntax of this option, check the "Video
12094 size" section in the ffmpeg-utils manual. Default value is "640x480".
12097 Set the initial scale value. Default value is 3.0.
12100 Set the initial x position. Must be a floating point value between
12101 -100 and 100. Default value is -0.743643887037158704752191506114774.
12104 Set the initial y position. Must be a floating point value between
12105 -100 and 100. Default value is -0.131825904205311970493132056385139.
12110 Generate various test patterns, as generated by the MPlayer test filter.
12112 The size of the generated video is fixed, and is 256x256.
12113 This source is useful in particular for testing encoding features.
12115 This source accepts the following options:
12120 Specify the frame rate of the sourced video, as the number of frames
12121 generated per second. It has to be a string in the format
12122 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
12123 number or a valid video frame rate abbreviation. The default value is
12127 Set the duration of the sourced video. See
12128 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12129 for the accepted syntax.
12131 If not specified, or the expressed duration is negative, the video is
12132 supposed to be generated forever.
12136 Set the number or the name of the test to perform. Supported tests are:
12152 Default value is "all", which will cycle through the list of all tests.
12157 mptestsrc=t=dc_luma
12160 will generate a "dc_luma" test pattern.
12162 @section frei0r_src
12164 Provide a frei0r source.
12166 To enable compilation of this filter you need to install the frei0r
12167 header and configure FFmpeg with @code{--enable-frei0r}.
12169 This source accepts the following parameters:
12174 The size of the video to generate. For the syntax of this option, check the
12175 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12178 The framerate of the generated video. It may be a string of the form
12179 @var{num}/@var{den} or a frame rate abbreviation.
12182 The name to the frei0r source to load. For more information regarding frei0r and
12183 how to set the parameters, read the @ref{frei0r} section in the video filters
12186 @item filter_params
12187 A '|'-separated list of parameters to pass to the frei0r source.
12191 For example, to generate a frei0r partik0l source with size 200x200
12192 and frame rate 10 which is overlaid on the overlay filter main input:
12194 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
12199 Generate a life pattern.
12201 This source is based on a generalization of John Conway's life game.
12203 The sourced input represents a life grid, each pixel represents a cell
12204 which can be in one of two possible states, alive or dead. Every cell
12205 interacts with its eight neighbours, which are the cells that are
12206 horizontally, vertically, or diagonally adjacent.
12208 At each interaction the grid evolves according to the adopted rule,
12209 which specifies the number of neighbor alive cells which will make a
12210 cell stay alive or born. The @option{rule} option allows one to specify
12213 This source accepts the following options:
12217 Set the file from which to read the initial grid state. In the file,
12218 each non-whitespace character is considered an alive cell, and newline
12219 is used to delimit the end of each row.
12221 If this option is not specified, the initial grid is generated
12225 Set the video rate, that is the number of frames generated per second.
12228 @item random_fill_ratio, ratio
12229 Set the random fill ratio for the initial random grid. It is a
12230 floating point number value ranging from 0 to 1, defaults to 1/PHI.
12231 It is ignored when a file is specified.
12233 @item random_seed, seed
12234 Set the seed for filling the initial random grid, must be an integer
12235 included between 0 and UINT32_MAX. If not specified, or if explicitly
12236 set to -1, the filter will try to use a good random seed on a best
12242 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
12243 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
12244 @var{NS} specifies the number of alive neighbor cells which make a
12245 live cell stay alive, and @var{NB} the number of alive neighbor cells
12246 which make a dead cell to become alive (i.e. to "born").
12247 "s" and "b" can be used in place of "S" and "B", respectively.
12249 Alternatively a rule can be specified by an 18-bits integer. The 9
12250 high order bits are used to encode the next cell state if it is alive
12251 for each number of neighbor alive cells, the low order bits specify
12252 the rule for "borning" new cells. Higher order bits encode for an
12253 higher number of neighbor cells.
12254 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
12255 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
12257 Default value is "S23/B3", which is the original Conway's game of life
12258 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
12259 cells, and will born a new cell if there are three alive cells around
12263 Set the size of the output video. For the syntax of this option, check the
12264 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12266 If @option{filename} is specified, the size is set by default to the
12267 same size of the input file. If @option{size} is set, it must contain
12268 the size specified in the input file, and the initial grid defined in
12269 that file is centered in the larger resulting area.
12271 If a filename is not specified, the size value defaults to "320x240"
12272 (used for a randomly generated initial grid).
12275 If set to 1, stitch the left and right grid edges together, and the
12276 top and bottom edges also. Defaults to 1.
12279 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
12280 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
12281 value from 0 to 255.
12284 Set the color of living (or new born) cells.
12287 Set the color of dead cells. If @option{mold} is set, this is the first color
12288 used to represent a dead cell.
12291 Set mold color, for definitely dead and moldy cells.
12293 For the syntax of these 3 color options, check the "Color" section in the
12294 ffmpeg-utils manual.
12297 @subsection Examples
12301 Read a grid from @file{pattern}, and center it on a grid of size
12304 life=f=pattern:s=300x300
12308 Generate a random grid of size 200x200, with a fill ratio of 2/3:
12310 life=ratio=2/3:s=200x200
12314 Specify a custom rule for evolving a randomly generated grid:
12320 Full example with slow death effect (mold) using @command{ffplay}:
12322 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
12329 @anchor{haldclutsrc}
12331 @anchor{rgbtestsrc}
12333 @anchor{smptehdbars}
12335 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
12337 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
12339 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
12341 The @code{color} source provides an uniformly colored input.
12343 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
12344 @ref{haldclut} filter.
12346 The @code{nullsrc} source returns unprocessed video frames. It is
12347 mainly useful to be employed in analysis / debugging tools, or as the
12348 source for filters which ignore the input data.
12350 The @code{rgbtestsrc} source generates an RGB test pattern useful for
12351 detecting RGB vs BGR issues. You should see a red, green and blue
12352 stripe from top to bottom.
12354 The @code{smptebars} source generates a color bars pattern, based on
12355 the SMPTE Engineering Guideline EG 1-1990.
12357 The @code{smptehdbars} source generates a color bars pattern, based on
12358 the SMPTE RP 219-2002.
12360 The @code{testsrc} source generates a test video pattern, showing a
12361 color pattern, a scrolling gradient and a timestamp. This is mainly
12362 intended for testing purposes.
12364 The sources accept the following parameters:
12369 Specify the color of the source, only available in the @code{color}
12370 source. For the syntax of this option, check the "Color" section in the
12371 ffmpeg-utils manual.
12374 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
12375 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
12376 pixels to be used as identity matrix for 3D lookup tables. Each component is
12377 coded on a @code{1/(N*N)} scale.
12380 Specify the size of the sourced video. For the syntax of this option, check the
12381 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12382 The default value is @code{320x240}.
12384 This option is not available with the @code{haldclutsrc} filter.
12387 Specify the frame rate of the sourced video, as the number of frames
12388 generated per second. It has to be a string in the format
12389 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
12390 number or a valid video frame rate abbreviation. The default value is
12394 Set the sample aspect ratio of the sourced video.
12397 Set the duration of the sourced video. See
12398 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12399 for the accepted syntax.
12401 If not specified, or the expressed duration is negative, the video is
12402 supposed to be generated forever.
12405 Set the number of decimals to show in the timestamp, only available in the
12406 @code{testsrc} source.
12408 The displayed timestamp value will correspond to the original
12409 timestamp value multiplied by the power of 10 of the specified
12410 value. Default value is 0.
12413 For example the following:
12415 testsrc=duration=5.3:size=qcif:rate=10
12418 will generate a video with a duration of 5.3 seconds, with size
12419 176x144 and a frame rate of 10 frames per second.
12421 The following graph description will generate a red source
12422 with an opacity of 0.2, with size "qcif" and a frame rate of 10
12425 color=c=red@@0.2:s=qcif:r=10
12428 If the input content is to be ignored, @code{nullsrc} can be used. The
12429 following command generates noise in the luminance plane by employing
12430 the @code{geq} filter:
12432 nullsrc=s=256x256, geq=random(1)*255:128:128
12435 @subsection Commands
12437 The @code{color} source supports the following commands:
12441 Set the color of the created image. Accepts the same syntax of the
12442 corresponding @option{color} option.
12445 @c man end VIDEO SOURCES
12447 @chapter Video Sinks
12448 @c man begin VIDEO SINKS
12450 Below is a description of the currently available video sinks.
12452 @section buffersink
12454 Buffer video frames, and make them available to the end of the filter
12457 This sink is mainly intended for programmatic use, in particular
12458 through the interface defined in @file{libavfilter/buffersink.h}
12459 or the options system.
12461 It accepts a pointer to an AVBufferSinkContext structure, which
12462 defines the incoming buffers' formats, to be passed as the opaque
12463 parameter to @code{avfilter_init_filter} for initialization.
12467 Null video sink: do absolutely nothing with the input video. It is
12468 mainly useful as a template and for use in analysis / debugging
12471 @c man end VIDEO SINKS
12473 @chapter Multimedia Filters
12474 @c man begin MULTIMEDIA FILTERS
12476 Below is a description of the currently available multimedia filters.
12478 @section aphasemeter
12480 Convert input audio to a video output, displaying the audio phase.
12482 The filter accepts the following options:
12486 Set the output frame rate. Default value is @code{25}.
12489 Set the video size for the output. For the syntax of this option, check the
12490 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12491 Default value is @code{800x400}.
12496 Specify the red, green, blue contrast. Default values are @code{2},
12497 @code{7} and @code{1}.
12498 Allowed range is @code{[0, 255]}.
12501 Set color which will be used for drawing median phase. If color is
12502 @code{none} which is default, no median phase value will be drawn.
12505 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
12506 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
12507 The @code{-1} means left and right channels are completely out of phase and
12508 @code{1} means channels are in phase.
12510 @section avectorscope
12512 Convert input audio to a video output, representing the audio vector
12515 The filter is used to measure the difference between channels of stereo
12516 audio stream. A monoaural signal, consisting of identical left and right
12517 signal, results in straight vertical line. Any stereo separation is visible
12518 as a deviation from this line, creating a Lissajous figure.
12519 If the straight (or deviation from it) but horizontal line appears this
12520 indicates that the left and right channels are out of phase.
12522 The filter accepts the following options:
12526 Set the vectorscope mode.
12528 Available values are:
12531 Lissajous rotated by 45 degrees.
12534 Same as above but not rotated.
12537 Shape resembling half of circle.
12540 Default value is @samp{lissajous}.
12543 Set the video size for the output. For the syntax of this option, check the
12544 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12545 Default value is @code{400x400}.
12548 Set the output frame rate. Default value is @code{25}.
12554 Specify the red, green, blue and alpha contrast. Default values are @code{40},
12555 @code{160}, @code{80} and @code{255}.
12556 Allowed range is @code{[0, 255]}.
12562 Specify the red, green, blue and alpha fade. Default values are @code{15},
12563 @code{10}, @code{5} and @code{5}.
12564 Allowed range is @code{[0, 255]}.
12567 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
12570 @subsection Examples
12574 Complete example using @command{ffplay}:
12576 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
12577 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
12583 Concatenate audio and video streams, joining them together one after the
12586 The filter works on segments of synchronized video and audio streams. All
12587 segments must have the same number of streams of each type, and that will
12588 also be the number of streams at output.
12590 The filter accepts the following options:
12595 Set the number of segments. Default is 2.
12598 Set the number of output video streams, that is also the number of video
12599 streams in each segment. Default is 1.
12602 Set the number of output audio streams, that is also the number of audio
12603 streams in each segment. Default is 0.
12606 Activate unsafe mode: do not fail if segments have a different format.
12610 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
12611 @var{a} audio outputs.
12613 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
12614 segment, in the same order as the outputs, then the inputs for the second
12617 Related streams do not always have exactly the same duration, for various
12618 reasons including codec frame size or sloppy authoring. For that reason,
12619 related synchronized streams (e.g. a video and its audio track) should be
12620 concatenated at once. The concat filter will use the duration of the longest
12621 stream in each segment (except the last one), and if necessary pad shorter
12622 audio streams with silence.
12624 For this filter to work correctly, all segments must start at timestamp 0.
12626 All corresponding streams must have the same parameters in all segments; the
12627 filtering system will automatically select a common pixel format for video
12628 streams, and a common sample format, sample rate and channel layout for
12629 audio streams, but other settings, such as resolution, must be converted
12630 explicitly by the user.
12632 Different frame rates are acceptable but will result in variable frame rate
12633 at output; be sure to configure the output file to handle it.
12635 @subsection Examples
12639 Concatenate an opening, an episode and an ending, all in bilingual version
12640 (video in stream 0, audio in streams 1 and 2):
12642 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
12643 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
12644 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
12645 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
12649 Concatenate two parts, handling audio and video separately, using the
12650 (a)movie sources, and adjusting the resolution:
12652 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
12653 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
12654 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
12656 Note that a desync will happen at the stitch if the audio and video streams
12657 do not have exactly the same duration in the first file.
12664 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
12665 it unchanged. By default, it logs a message at a frequency of 10Hz with the
12666 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
12667 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
12669 The filter also has a video output (see the @var{video} option) with a real
12670 time graph to observe the loudness evolution. The graphic contains the logged
12671 message mentioned above, so it is not printed anymore when this option is set,
12672 unless the verbose logging is set. The main graphing area contains the
12673 short-term loudness (3 seconds of analysis), and the gauge on the right is for
12674 the momentary loudness (400 milliseconds).
12676 More information about the Loudness Recommendation EBU R128 on
12677 @url{http://tech.ebu.ch/loudness}.
12679 The filter accepts the following options:
12684 Activate the video output. The audio stream is passed unchanged whether this
12685 option is set or no. The video stream will be the first output stream if
12686 activated. Default is @code{0}.
12689 Set the video size. This option is for video only. For the syntax of this
12691 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12692 Default and minimum resolution is @code{640x480}.
12695 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
12696 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
12697 other integer value between this range is allowed.
12700 Set metadata injection. If set to @code{1}, the audio input will be segmented
12701 into 100ms output frames, each of them containing various loudness information
12702 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
12704 Default is @code{0}.
12707 Force the frame logging level.
12709 Available values are:
12712 information logging level
12714 verbose logging level
12717 By default, the logging level is set to @var{info}. If the @option{video} or
12718 the @option{metadata} options are set, it switches to @var{verbose}.
12723 Available modes can be cumulated (the option is a @code{flag} type). Possible
12727 Disable any peak mode (default).
12729 Enable sample-peak mode.
12731 Simple peak mode looking for the higher sample value. It logs a message
12732 for sample-peak (identified by @code{SPK}).
12734 Enable true-peak mode.
12736 If enabled, the peak lookup is done on an over-sampled version of the input
12737 stream for better peak accuracy. It logs a message for true-peak.
12738 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
12739 This mode requires a build with @code{libswresample}.
12743 Treat mono input files as "dual mono". If a mono file is intended for playback
12744 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
12745 If set to @code{true}, this option will compensate for this effect.
12746 Multi-channel input files are not effected by this option.
12749 Set a specific pan law to be used for the measurement of dual mono files.
12750 This parameter is optional, and has a default value of -3.01dB.
12753 @subsection Examples
12757 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
12759 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
12763 Run an analysis with @command{ffmpeg}:
12765 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
12769 @section interleave, ainterleave
12771 Temporally interleave frames from several inputs.
12773 @code{interleave} works with video inputs, @code{ainterleave} with audio.
12775 These filters read frames from several inputs and send the oldest
12776 queued frame to the output.
12778 Input streams must have a well defined, monotonically increasing frame
12781 In order to submit one frame to output, these filters need to enqueue
12782 at least one frame for each input, so they cannot work in case one
12783 input is not yet terminated and will not receive incoming frames.
12785 For example consider the case when one input is a @code{select} filter
12786 which always drop input frames. The @code{interleave} filter will keep
12787 reading from that input, but it will never be able to send new frames
12788 to output until the input will send an end-of-stream signal.
12790 Also, depending on inputs synchronization, the filters will drop
12791 frames in case one input receives more frames than the other ones, and
12792 the queue is already filled.
12794 These filters accept the following options:
12798 Set the number of different inputs, it is 2 by default.
12801 @subsection Examples
12805 Interleave frames belonging to different streams using @command{ffmpeg}:
12807 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
12811 Add flickering blur effect:
12813 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
12817 @section perms, aperms
12819 Set read/write permissions for the output frames.
12821 These filters are mainly aimed at developers to test direct path in the
12822 following filter in the filtergraph.
12824 The filters accept the following options:
12828 Select the permissions mode.
12830 It accepts the following values:
12833 Do nothing. This is the default.
12835 Set all the output frames read-only.
12837 Set all the output frames directly writable.
12839 Make the frame read-only if writable, and writable if read-only.
12841 Set each output frame read-only or writable randomly.
12845 Set the seed for the @var{random} mode, must be an integer included between
12846 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
12847 @code{-1}, the filter will try to use a good random seed on a best effort
12851 Note: in case of auto-inserted filter between the permission filter and the
12852 following one, the permission might not be received as expected in that
12853 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
12854 perms/aperms filter can avoid this problem.
12856 @section select, aselect
12858 Select frames to pass in output.
12860 This filter accepts the following options:
12865 Set expression, which is evaluated for each input frame.
12867 If the expression is evaluated to zero, the frame is discarded.
12869 If the evaluation result is negative or NaN, the frame is sent to the
12870 first output; otherwise it is sent to the output with index
12871 @code{ceil(val)-1}, assuming that the input index starts from 0.
12873 For example a value of @code{1.2} corresponds to the output with index
12874 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
12877 Set the number of outputs. The output to which to send the selected
12878 frame is based on the result of the evaluation. Default value is 1.
12881 The expression can contain the following constants:
12885 The (sequential) number of the filtered frame, starting from 0.
12888 The (sequential) number of the selected frame, starting from 0.
12890 @item prev_selected_n
12891 The sequential number of the last selected frame. It's NAN if undefined.
12894 The timebase of the input timestamps.
12897 The PTS (Presentation TimeStamp) of the filtered video frame,
12898 expressed in @var{TB} units. It's NAN if undefined.
12901 The PTS of the filtered video frame,
12902 expressed in seconds. It's NAN if undefined.
12905 The PTS of the previously filtered video frame. It's NAN if undefined.
12907 @item prev_selected_pts
12908 The PTS of the last previously filtered video frame. It's NAN if undefined.
12910 @item prev_selected_t
12911 The PTS of the last previously selected video frame. It's NAN if undefined.
12914 The PTS of the first video frame in the video. It's NAN if undefined.
12917 The time of the first video frame in the video. It's NAN if undefined.
12919 @item pict_type @emph{(video only)}
12920 The type of the filtered frame. It can assume one of the following
12932 @item interlace_type @emph{(video only)}
12933 The frame interlace type. It can assume one of the following values:
12936 The frame is progressive (not interlaced).
12938 The frame is top-field-first.
12940 The frame is bottom-field-first.
12943 @item consumed_sample_n @emph{(audio only)}
12944 the number of selected samples before the current frame
12946 @item samples_n @emph{(audio only)}
12947 the number of samples in the current frame
12949 @item sample_rate @emph{(audio only)}
12950 the input sample rate
12953 This is 1 if the filtered frame is a key-frame, 0 otherwise.
12956 the position in the file of the filtered frame, -1 if the information
12957 is not available (e.g. for synthetic video)
12959 @item scene @emph{(video only)}
12960 value between 0 and 1 to indicate a new scene; a low value reflects a low
12961 probability for the current frame to introduce a new scene, while a higher
12962 value means the current frame is more likely to be one (see the example below)
12966 The default value of the select expression is "1".
12968 @subsection Examples
12972 Select all frames in input:
12977 The example above is the same as:
12989 Select only I-frames:
12991 select='eq(pict_type\,I)'
12995 Select one frame every 100:
12997 select='not(mod(n\,100))'
13001 Select only frames contained in the 10-20 time interval:
13003 select=between(t\,10\,20)
13007 Select only I frames contained in the 10-20 time interval:
13009 select=between(t\,10\,20)*eq(pict_type\,I)
13013 Select frames with a minimum distance of 10 seconds:
13015 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
13019 Use aselect to select only audio frames with samples number > 100:
13021 aselect='gt(samples_n\,100)'
13025 Create a mosaic of the first scenes:
13027 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
13030 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
13034 Send even and odd frames to separate outputs, and compose them:
13036 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
13040 @section sendcmd, asendcmd
13042 Send commands to filters in the filtergraph.
13044 These filters read commands to be sent to other filters in the
13047 @code{sendcmd} must be inserted between two video filters,
13048 @code{asendcmd} must be inserted between two audio filters, but apart
13049 from that they act the same way.
13051 The specification of commands can be provided in the filter arguments
13052 with the @var{commands} option, or in a file specified by the
13053 @var{filename} option.
13055 These filters accept the following options:
13058 Set the commands to be read and sent to the other filters.
13060 Set the filename of the commands to be read and sent to the other
13064 @subsection Commands syntax
13066 A commands description consists of a sequence of interval
13067 specifications, comprising a list of commands to be executed when a
13068 particular event related to that interval occurs. The occurring event
13069 is typically the current frame time entering or leaving a given time
13072 An interval is specified by the following syntax:
13074 @var{START}[-@var{END}] @var{COMMANDS};
13077 The time interval is specified by the @var{START} and @var{END} times.
13078 @var{END} is optional and defaults to the maximum time.
13080 The current frame time is considered within the specified interval if
13081 it is included in the interval [@var{START}, @var{END}), that is when
13082 the time is greater or equal to @var{START} and is lesser than
13085 @var{COMMANDS} consists of a sequence of one or more command
13086 specifications, separated by ",", relating to that interval. The
13087 syntax of a command specification is given by:
13089 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
13092 @var{FLAGS} is optional and specifies the type of events relating to
13093 the time interval which enable sending the specified command, and must
13094 be a non-null sequence of identifier flags separated by "+" or "|" and
13095 enclosed between "[" and "]".
13097 The following flags are recognized:
13100 The command is sent when the current frame timestamp enters the
13101 specified interval. In other words, the command is sent when the
13102 previous frame timestamp was not in the given interval, and the
13106 The command is sent when the current frame timestamp leaves the
13107 specified interval. In other words, the command is sent when the
13108 previous frame timestamp was in the given interval, and the
13112 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
13115 @var{TARGET} specifies the target of the command, usually the name of
13116 the filter class or a specific filter instance name.
13118 @var{COMMAND} specifies the name of the command for the target filter.
13120 @var{ARG} is optional and specifies the optional list of argument for
13121 the given @var{COMMAND}.
13123 Between one interval specification and another, whitespaces, or
13124 sequences of characters starting with @code{#} until the end of line,
13125 are ignored and can be used to annotate comments.
13127 A simplified BNF description of the commands specification syntax
13130 @var{COMMAND_FLAG} ::= "enter" | "leave"
13131 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
13132 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
13133 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
13134 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
13135 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
13138 @subsection Examples
13142 Specify audio tempo change at second 4:
13144 asendcmd=c='4.0 atempo tempo 1.5',atempo
13148 Specify a list of drawtext and hue commands in a file.
13150 # show text in the interval 5-10
13151 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
13152 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
13154 # desaturate the image in the interval 15-20
13155 15.0-20.0 [enter] hue s 0,
13156 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
13158 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
13160 # apply an exponential saturation fade-out effect, starting from time 25
13161 25 [enter] hue s exp(25-t)
13164 A filtergraph allowing to read and process the above command list
13165 stored in a file @file{test.cmd}, can be specified with:
13167 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
13172 @section setpts, asetpts
13174 Change the PTS (presentation timestamp) of the input frames.
13176 @code{setpts} works on video frames, @code{asetpts} on audio frames.
13178 This filter accepts the following options:
13183 The expression which is evaluated for each frame to construct its timestamp.
13187 The expression is evaluated through the eval API and can contain the following
13192 frame rate, only defined for constant frame-rate video
13195 The presentation timestamp in input
13198 The count of the input frame for video or the number of consumed samples,
13199 not including the current frame for audio, starting from 0.
13201 @item NB_CONSUMED_SAMPLES
13202 The number of consumed samples, not including the current frame (only
13205 @item NB_SAMPLES, S
13206 The number of samples in the current frame (only audio)
13208 @item SAMPLE_RATE, SR
13209 The audio sample rate.
13212 The PTS of the first frame.
13215 the time in seconds of the first frame
13218 State whether the current frame is interlaced.
13221 the time in seconds of the current frame
13224 original position in the file of the frame, or undefined if undefined
13225 for the current frame
13228 The previous input PTS.
13231 previous input time in seconds
13234 The previous output PTS.
13237 previous output time in seconds
13240 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
13244 The wallclock (RTC) time at the start of the movie in microseconds.
13247 The timebase of the input timestamps.
13251 @subsection Examples
13255 Start counting PTS from zero
13257 setpts=PTS-STARTPTS
13261 Apply fast motion effect:
13267 Apply slow motion effect:
13273 Set fixed rate of 25 frames per second:
13279 Set fixed rate 25 fps with some jitter:
13281 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
13285 Apply an offset of 10 seconds to the input PTS:
13291 Generate timestamps from a "live source" and rebase onto the current timebase:
13293 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
13297 Generate timestamps by counting samples:
13304 @section settb, asettb
13306 Set the timebase to use for the output frames timestamps.
13307 It is mainly useful for testing timebase configuration.
13309 It accepts the following parameters:
13314 The expression which is evaluated into the output timebase.
13318 The value for @option{tb} is an arithmetic expression representing a
13319 rational. The expression can contain the constants "AVTB" (the default
13320 timebase), "intb" (the input timebase) and "sr" (the sample rate,
13321 audio only). Default value is "intb".
13323 @subsection Examples
13327 Set the timebase to 1/25:
13333 Set the timebase to 1/10:
13339 Set the timebase to 1001/1000:
13345 Set the timebase to 2*intb:
13351 Set the default timebase value:
13358 Convert input audio to a video output representing
13359 frequency spectrum logarithmically (using constant Q transform with
13360 Brown-Puckette algorithm), with musical tone scale, from E0 to D#10 (10 octaves).
13362 The filter accepts the following options:
13366 Specify transform volume (multiplier) expression. The expression can contain
13369 @item frequency, freq, f
13370 the frequency where transform is evaluated
13371 @item timeclamp, tc
13372 value of timeclamp option
13376 @item a_weighting(f)
13377 A-weighting of equal loudness
13378 @item b_weighting(f)
13379 B-weighting of equal loudness
13380 @item c_weighting(f)
13381 C-weighting of equal loudness
13383 Default value is @code{16}.
13386 Specify transform length expression. The expression can contain variables:
13388 @item frequency, freq, f
13389 the frequency where transform is evaluated
13390 @item timeclamp, tc
13391 value of timeclamp option
13393 Default value is @code{384/f*tc/(384/f+tc)}.
13396 Specify the transform timeclamp. At low frequency, there is trade-off between
13397 accuracy in time domain and frequency domain. If timeclamp is lower,
13398 event in time domain is represented more accurately (such as fast bass drum),
13399 otherwise event in frequency domain is represented more accurately
13400 (such as bass guitar). Acceptable value is [0.1, 1.0]. Default value is @code{0.17}.
13403 Specify the transform coeffclamp. If coeffclamp is lower, transform is
13404 more accurate, otherwise transform is faster. Acceptable value is [0.1, 10.0].
13405 Default value is @code{1.0}.
13408 Specify gamma. Lower gamma makes the spectrum more contrast, higher gamma
13409 makes the spectrum having more range. Acceptable value is [1.0, 7.0].
13410 Default value is @code{3.0}.
13413 Specify gamma of bargraph. Acceptable value is [1.0, 7.0].
13414 Default value is @code{1.0}.
13417 Specify font file for use with freetype. If not specified, use embedded font.
13420 Specify font color expression. This is arithmetic expression that should return
13421 integer value 0xRRGGBB. The expression can contain variables:
13423 @item frequency, freq, f
13424 the frequency where transform is evaluated
13425 @item timeclamp, tc
13426 value of timeclamp option
13431 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
13432 @item r(x), g(x), b(x)
13433 red, green, and blue value of intensity x
13435 Default value is @code{st(0, (midi(f)-59.5)/12);
13436 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
13437 r(1-ld(1)) + b(ld(1))}
13440 If set to 1 (the default), the video size is 1920x1080 (full HD),
13441 if set to 0, the video size is 960x540. Use this option to make CPU usage lower.
13444 Specify video fps. Default value is @code{25}.
13447 Specify number of transform per frame, so there are fps*count transforms
13448 per second. Note that audio data rate must be divisible by fps*count.
13449 Default value is @code{6}.
13453 @subsection Examples
13457 Playing audio while showing the spectrum:
13459 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
13463 Same as above, but with frame rate 30 fps:
13465 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
13469 Playing at 960x540 and lower CPU usage:
13471 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fullhd=0:count=3 [out0]'
13475 A1 and its harmonics: A1, A2, (near)E3, A3:
13477 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),
13478 asplit[a][out1]; [a] showcqt [out0]'
13482 Same as above, but with more accuracy in frequency domain (and slower):
13484 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),
13485 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
13489 B-weighting of equal loudness
13491 volume=16*b_weighting(f)
13497 tlength=100/f*tc/(100/f+tc)
13501 Custom fontcolor, C-note is colored green, others are colored blue
13503 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))'
13507 Custom gamma, now spectrum is linear to the amplitude.
13516 Convert input audio to video output representing the audio power spectrum.
13517 Audio amplitude is on Y-axis while frequency is on X-axis.
13519 The filter accepts the following options:
13523 Specify size of video. For the syntax of this option, check the
13524 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13525 Default is @code{1024x512}.
13529 This set how each frequency bin will be represented.
13531 It accepts the following values:
13537 Default is @code{bar}.
13540 Set amplitude scale.
13542 It accepts the following values:
13556 Default is @code{log}.
13559 Set frequency scale.
13561 It accepts the following values:
13570 Reverse logarithmic scale.
13572 Default is @code{lin}.
13577 It accepts the following values:
13593 Default is @code{w2048}
13596 Set windowing function.
13598 It accepts the following values:
13615 Default is @code{hanning}.
13618 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
13619 which means optimal overlap for selected window function will be picked.
13622 Set time averaging. Setting this to 0 will display current maximal peaks.
13623 Default is @code{1}, which means time averaging is disabled.
13626 Specify list of colors separated by space or by '|' which will be used to
13627 draw channel frequencies. Unrecognized or missing colors will be replaced
13631 @section showspectrum
13633 Convert input audio to a video output, representing the audio frequency
13636 The filter accepts the following options:
13640 Specify the video size for the output. For the syntax of this option, check the
13641 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13642 Default value is @code{640x512}.
13645 Specify how the spectrum should slide along the window.
13647 It accepts the following values:
13650 the samples start again on the left when they reach the right
13652 the samples scroll from right to left
13654 frames are only produced when the samples reach the right
13657 Default value is @code{replace}.
13660 Specify display mode.
13662 It accepts the following values:
13665 all channels are displayed in the same row
13667 all channels are displayed in separate rows
13670 Default value is @samp{combined}.
13673 Specify display color mode.
13675 It accepts the following values:
13678 each channel is displayed in a separate color
13680 each channel is is displayed using the same color scheme
13683 Default value is @samp{channel}.
13686 Specify scale used for calculating intensity color values.
13688 It accepts the following values:
13693 square root, default
13700 Default value is @samp{sqrt}.
13703 Set saturation modifier for displayed colors. Negative values provide
13704 alternative color scheme. @code{0} is no saturation at all.
13705 Saturation must be in [-10.0, 10.0] range.
13706 Default value is @code{1}.
13709 Set window function.
13711 It accepts the following values:
13714 No samples pre-processing (do not expect this to be faster)
13723 Default value is @code{hann}.
13726 The usage is very similar to the showwaves filter; see the examples in that
13729 @subsection Examples
13733 Large window with logarithmic color scaling:
13735 showspectrum=s=1280x480:scale=log
13739 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
13741 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
13742 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
13746 @section showvolume
13748 Convert input audio volume to a video output.
13750 The filter accepts the following options:
13757 Set border width, allowed range is [0, 5]. Default is 1.
13760 Set channel width, allowed range is [40, 1080]. Default is 400.
13763 Set channel height, allowed range is [1, 100]. Default is 20.
13766 Set fade, allowed range is [1, 255]. Default is 20.
13769 Set volume color expression.
13771 The expression can use the following variables:
13775 Current max volume of channel in dB.
13778 Current channel number, starting from 0.
13782 If set, displays channel names. Default is enabled.
13787 Convert input audio to a video output, representing the samples waves.
13789 The filter accepts the following options:
13793 Specify the video size for the output. For the syntax of this option, check the
13794 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13795 Default value is @code{600x240}.
13800 Available values are:
13803 Draw a point for each sample.
13806 Draw a vertical line for each sample.
13809 Draw a point for each sample and a line between them.
13812 Draw a centered vertical line for each sample.
13815 Default value is @code{point}.
13818 Set the number of samples which are printed on the same column. A
13819 larger value will decrease the frame rate. Must be a positive
13820 integer. This option can be set only if the value for @var{rate}
13821 is not explicitly specified.
13824 Set the (approximate) output frame rate. This is done by setting the
13825 option @var{n}. Default value is "25".
13827 @item split_channels
13828 Set if channels should be drawn separately or overlap. Default value is 0.
13832 @subsection Examples
13836 Output the input file audio and the corresponding video representation
13839 amovie=a.mp3,asplit[out0],showwaves[out1]
13843 Create a synthetic signal and show it with showwaves, forcing a
13844 frame rate of 30 frames per second:
13846 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
13850 @section showwavespic
13852 Convert input audio to a single video frame, representing the samples waves.
13854 The filter accepts the following options:
13858 Specify the video size for the output. For the syntax of this option, check the
13859 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13860 Default value is @code{600x240}.
13862 @item split_channels
13863 Set if channels should be drawn separately or overlap. Default value is 0.
13866 @subsection Examples
13870 Extract a channel split representation of the wave form of a whole audio track
13871 in a 1024x800 picture using @command{ffmpeg}:
13873 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
13877 @section split, asplit
13879 Split input into several identical outputs.
13881 @code{asplit} works with audio input, @code{split} with video.
13883 The filter accepts a single parameter which specifies the number of outputs. If
13884 unspecified, it defaults to 2.
13886 @subsection Examples
13890 Create two separate outputs from the same input:
13892 [in] split [out0][out1]
13896 To create 3 or more outputs, you need to specify the number of
13899 [in] asplit=3 [out0][out1][out2]
13903 Create two separate outputs from the same input, one cropped and
13906 [in] split [splitout1][splitout2];
13907 [splitout1] crop=100:100:0:0 [cropout];
13908 [splitout2] pad=200:200:100:100 [padout];
13912 Create 5 copies of the input audio with @command{ffmpeg}:
13914 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
13920 Receive commands sent through a libzmq client, and forward them to
13921 filters in the filtergraph.
13923 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
13924 must be inserted between two video filters, @code{azmq} between two
13927 To enable these filters you need to install the libzmq library and
13928 headers and configure FFmpeg with @code{--enable-libzmq}.
13930 For more information about libzmq see:
13931 @url{http://www.zeromq.org/}
13933 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
13934 receives messages sent through a network interface defined by the
13935 @option{bind_address} option.
13937 The received message must be in the form:
13939 @var{TARGET} @var{COMMAND} [@var{ARG}]
13942 @var{TARGET} specifies the target of the command, usually the name of
13943 the filter class or a specific filter instance name.
13945 @var{COMMAND} specifies the name of the command for the target filter.
13947 @var{ARG} is optional and specifies the optional argument list for the
13948 given @var{COMMAND}.
13950 Upon reception, the message is processed and the corresponding command
13951 is injected into the filtergraph. Depending on the result, the filter
13952 will send a reply to the client, adopting the format:
13954 @var{ERROR_CODE} @var{ERROR_REASON}
13958 @var{MESSAGE} is optional.
13960 @subsection Examples
13962 Look at @file{tools/zmqsend} for an example of a zmq client which can
13963 be used to send commands processed by these filters.
13965 Consider the following filtergraph generated by @command{ffplay}
13967 ffplay -dumpgraph 1 -f lavfi "
13968 color=s=100x100:c=red [l];
13969 color=s=100x100:c=blue [r];
13970 nullsrc=s=200x100, zmq [bg];
13971 [bg][l] overlay [bg+l];
13972 [bg+l][r] overlay=x=100 "
13975 To change the color of the left side of the video, the following
13976 command can be used:
13978 echo Parsed_color_0 c yellow | tools/zmqsend
13981 To change the right side:
13983 echo Parsed_color_1 c pink | tools/zmqsend
13986 @c man end MULTIMEDIA FILTERS
13988 @chapter Multimedia Sources
13989 @c man begin MULTIMEDIA SOURCES
13991 Below is a description of the currently available multimedia sources.
13995 This is the same as @ref{movie} source, except it selects an audio
14001 Read audio and/or video stream(s) from a movie container.
14003 It accepts the following parameters:
14007 The name of the resource to read (not necessarily a file; it can also be a
14008 device or a stream accessed through some protocol).
14010 @item format_name, f
14011 Specifies the format assumed for the movie to read, and can be either
14012 the name of a container or an input device. If not specified, the
14013 format is guessed from @var{movie_name} or by probing.
14015 @item seek_point, sp
14016 Specifies the seek point in seconds. The frames will be output
14017 starting from this seek point. The parameter is evaluated with
14018 @code{av_strtod}, so the numerical value may be suffixed by an IS
14019 postfix. The default value is "0".
14022 Specifies the streams to read. Several streams can be specified,
14023 separated by "+". The source will then have as many outputs, in the
14024 same order. The syntax is explained in the ``Stream specifiers''
14025 section in the ffmpeg manual. Two special names, "dv" and "da" specify
14026 respectively the default (best suited) video and audio stream. Default
14027 is "dv", or "da" if the filter is called as "amovie".
14029 @item stream_index, si
14030 Specifies the index of the video stream to read. If the value is -1,
14031 the most suitable video stream will be automatically selected. The default
14032 value is "-1". Deprecated. If the filter is called "amovie", it will select
14033 audio instead of video.
14036 Specifies how many times to read the stream in sequence.
14037 If the value is less than 1, the stream will be read again and again.
14038 Default value is "1".
14040 Note that when the movie is looped the source timestamps are not
14041 changed, so it will generate non monotonically increasing timestamps.
14044 It allows overlaying a second video on top of the main input of
14045 a filtergraph, as shown in this graph:
14047 input -----------> deltapts0 --> overlay --> output
14050 movie --> scale--> deltapts1 -------+
14052 @subsection Examples
14056 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
14057 on top of the input labelled "in":
14059 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
14060 [in] setpts=PTS-STARTPTS [main];
14061 [main][over] overlay=16:16 [out]
14065 Read from a video4linux2 device, and overlay it on top of the input
14068 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
14069 [in] setpts=PTS-STARTPTS [main];
14070 [main][over] overlay=16:16 [out]
14074 Read the first video stream and the audio stream with id 0x81 from
14075 dvd.vob; the video is connected to the pad named "video" and the audio is
14076 connected to the pad named "audio":
14078 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
14082 @c man end MULTIMEDIA SOURCES