1 @chapter Filtergraph description
2 @c man begin FILTERGRAPH DESCRIPTION
4 A filtergraph is a directed graph of connected filters. It can contain
5 cycles, and there can be multiple links between a pair of
6 filters. Each link has one input pad on one side connecting it to one
7 filter from which it takes its input, and one output pad on the other
8 side connecting it to the one filter accepting its output.
10 Each filter in a filtergraph is an instance of a filter class
11 registered in the application, which defines the features and the
12 number of input and output pads of the filter.
14 A filter with no input pads is called a "source", a filter with no
15 output pads is called a "sink".
17 @section Filtergraph syntax
19 A filtergraph can be represented using a textual representation, which
20 is recognized by the @code{-vf} and @code{-af} options of the ff*
21 tools, and by the @code{avfilter_graph_parse()} function defined in
22 @file{libavfilter/avfiltergraph.h}.
24 A filterchain consists of a sequence of connected filters, each one
25 connected to the previous one in the sequence. A filterchain is
26 represented by a list of ","-separated filter descriptions.
28 A filtergraph consists of a sequence of filterchains. A sequence of
29 filterchains is represented by a list of ";"-separated filterchain
32 A filter is represented by a string of the form:
33 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
35 @var{filter_name} is the name of the filter class of which the
36 described filter is an instance of, and has to be the name of one of
37 the filter classes registered in the program.
38 The name of the filter class is optionally followed by a string
41 @var{arguments} is a string which contains the parameters used to
42 initialize the filter instance, and are described in the filter
45 The list of arguments can be quoted using the character "'" as initial
46 and ending mark, and the character '\' for escaping the characters
47 within the quoted text; otherwise the argument string is considered
48 terminated when the next special character (belonging to the set
49 "[]=;,") is encountered.
51 The name and arguments of the filter are optionally preceded and
52 followed by a list of link labels.
53 A link label allows to name a link and associate it to a filter output
54 or input pad. The preceding labels @var{in_link_1}
55 ... @var{in_link_N}, are associated to the filter input pads,
56 the following labels @var{out_link_1} ... @var{out_link_M}, are
57 associated to the output pads.
59 When two link labels with the same name are found in the
60 filtergraph, a link between the corresponding input and output pad is
63 If an output pad is not labelled, it is linked by default to the first
64 unlabelled input pad of the next filter in the filterchain.
65 For example in the filterchain:
67 nullsrc, split[L1], [L2]overlay, nullsink
69 the split filter instance has two output pads, and the overlay filter
70 instance two input pads. The first output pad of split is labelled
71 "L1", the first input pad of overlay is labelled "L2", and the second
72 output pad of split is linked to the second input pad of overlay,
73 which are both unlabelled.
75 In a complete filterchain all the unlabelled filter input and output
76 pads must be connected. A filtergraph is considered valid if all the
77 filter input and output pads of all the filterchains are connected.
79 Follows a BNF description for the filtergraph syntax:
81 @var{NAME} ::= sequence of alphanumeric characters and '_'
82 @var{LINKLABEL} ::= "[" @var{NAME} "]"
83 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
84 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
85 @var{FILTER} ::= [@var{LINKNAMES}] @var{NAME} ["=" @var{ARGUMENTS}] [@var{LINKNAMES}]
86 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
87 @var{FILTERGRAPH} ::= @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
90 @c man end FILTERGRAPH DESCRIPTION
92 @chapter Audio Filters
93 @c man begin AUDIO FILTERS
95 When you configure your FFmpeg build, you can disable any of the
96 existing filters using --disable-filters.
97 The configure output will show the audio filters included in your
100 Below is a description of the currently available audio filters.
104 Convert the input audio to one of the specified formats. The framework will
105 negotiate the most appropriate format to minimize conversions.
107 The filter accepts three lists of formats, separated by ":", in the form:
108 "@var{sample_formats}:@var{channel_layouts}:@var{packing_formats}".
110 Elements in each list are separated by "," which has to be escaped in the
111 filtergraph specification.
113 The special parameter "all", in place of a list of elements, signifies all
116 Some examples follow:
118 aformat=u8\\,s16:mono:packed
120 aformat=s16:mono\\,stereo:all
125 Pass the audio source unchanged to the output.
129 Resample the input audio to the specified sample rate.
131 The filter accepts exactly one parameter, the output sample rate. If not
132 specified then the filter will automatically convert between its input
133 and output sample rates.
135 For example, to resample the input audio to 44100Hz:
142 Show a line containing various information for each input audio frame.
143 The input audio is not modified.
145 The shown line contains a sequence of key/value pairs of the form
146 @var{key}:@var{value}.
148 A description of each shown parameter follows:
152 sequential number of the input frame, starting from 0
155 presentation TimeStamp of the input frame, expressed as a number of
156 time base units. The time base unit depends on the filter input pad, and
157 is usually 1/@var{sample_rate}.
160 presentation TimeStamp of the input frame, expressed as a number of
164 position of the frame in the input stream, -1 if this information in
165 unavailable and/or meanigless (for example in case of synthetic audio)
171 channel layout description
174 number of samples (per each channel) contained in the filtered frame
177 sample rate for the audio frame
180 if the packing format is planar, 0 if packed
183 Adler-32 checksum of all the planes of the input frame
186 Adler-32 checksum for each input frame plane, expressed in the form
187 "[@var{c0} @var{c1} @var{c2} @var{c3} @var{c4} @var{c5} @var{c6} @var{c7}]"
190 @c man end AUDIO FILTERS
192 @chapter Audio Sources
193 @c man begin AUDIO SOURCES
195 Below is a description of the currently available audio sources.
199 Buffer audio frames, and make them available to the filter chain.
201 This source is mainly intended for a programmatic use, in particular
202 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
204 It accepts the following mandatory parameters:
205 @var{sample_rate}:@var{sample_fmt}:@var{channel_layout}:@var{packing}
210 The sample rate of the incoming audio buffers.
213 The sample format of the incoming audio buffers.
214 Either a sample format name or its corresponging integer representation from
215 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
218 The channel layout of the incoming audio buffers.
219 Either a channel layout name from channel_layout_map in
220 @file{libavutil/audioconvert.c} or its corresponding integer representation
221 from the AV_CH_LAYOUT_* macros in @file{libavutil/audioconvert.h}
224 Either "packed" or "planar", or their integer representation: 0 or 1
231 abuffer=44100:s16:stereo:planar
234 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
235 Since the sample format with name "s16" corresponds to the number
236 1 and the "stereo" channel layout corresponds to the value 3, this is
244 Read an audio stream from a movie container.
246 It accepts the syntax: @var{movie_name}[:@var{options}] where
247 @var{movie_name} is the name of the resource to read (not necessarily
248 a file but also a device or a stream accessed through some protocol),
249 and @var{options} is an optional sequence of @var{key}=@var{value}
250 pairs, separated by ":".
252 The description of the accepted options follows.
257 Specify the format assumed for the movie to read, and can be either
258 the name of a container or an input device. If not specified the
259 format is guessed from @var{movie_name} or by probing.
262 Specify the seek point in seconds, the frames will be output
263 starting from this seek point, the parameter is evaluated with
264 @code{av_strtod} so the numerical value may be suffixed by an IS
265 postfix. Default value is "0".
267 @item stream_index, si
268 Specify the index of the audio stream to read. If the value is -1,
269 the best suited audio stream will be automatically selected. Default
276 Null audio source, never return audio frames. It is mainly useful as a
277 template and to be employed in analysis / debugging tools.
279 It accepts as optional parameter a string of the form
280 @var{sample_rate}:@var{channel_layout}.
282 @var{sample_rate} specify the sample rate, and defaults to 44100.
284 @var{channel_layout} specify the channel layout, and can be either an
285 integer or a string representing a channel layout. The default value
286 of @var{channel_layout} is 3, which corresponds to CH_LAYOUT_STEREO.
288 Check the channel_layout_map definition in
289 @file{libavcodec/audioconvert.c} for the mapping between strings and
290 channel layout values.
292 Follow some examples:
294 # set the sample rate to 48000 Hz and the channel layout to CH_LAYOUT_MONO.
301 @c man end AUDIO SOURCES
304 @c man begin AUDIO SINKS
306 Below is a description of the currently available audio sinks.
310 Buffer audio frames, and make them available to the end of filter chain.
312 This sink is mainly intended for programmatic use, in particular
313 through the interface defined in @file{libavfilter/buffersink.h}.
315 It requires a pointer to an AVABufferSinkContext structure, which
316 defines the incoming buffers' formats, to be passed as the opaque
317 parameter to @code{avfilter_init_filter} for initialization.
321 Null audio sink, do absolutely nothing with the input audio. It is
322 mainly useful as a template and to be employed in analysis / debugging
325 @c man end AUDIO SINKS
327 @chapter Video Filters
328 @c man begin VIDEO FILTERS
330 When you configure your FFmpeg build, you can disable any of the
331 existing filters using --disable-filters.
332 The configure output will show the video filters included in your
335 Below is a description of the currently available video filters.
339 Detect frames that are (almost) completely black. Can be useful to
340 detect chapter transitions or commercials. Output lines consist of
341 the frame number of the detected frame, the percentage of blackness,
342 the position in the file if known or -1 and the timestamp in seconds.
344 In order to display the output lines, you need to set the loglevel at
345 least to the AV_LOG_INFO value.
347 The filter accepts the syntax:
349 blackframe[=@var{amount}:[@var{threshold}]]
352 @var{amount} is the percentage of the pixels that have to be below the
353 threshold, and defaults to 98.
355 @var{threshold} is the threshold below which a pixel value is
356 considered black, and defaults to 32.
360 Apply boxblur algorithm to the input video.
362 This filter accepts the parameters:
363 @var{luma_radius}:@var{luma_power}:@var{chroma_radius}:@var{chroma_power}:@var{alpha_radius}:@var{alpha_power}
365 Chroma and alpha parameters are optional, if not specified they default
366 to the corresponding values set for @var{luma_radius} and
369 @var{luma_radius}, @var{chroma_radius}, and @var{alpha_radius} represent
370 the radius in pixels of the box used for blurring the corresponding
371 input plane. They are expressions, and can contain the following
375 the input width and heigth in pixels
378 the input chroma image width and height in pixels
381 horizontal and vertical chroma subsample values. For example for the
382 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
385 The radius must be a non-negative number, and must be not greater than
386 the value of the expression @code{min(w,h)/2} for the luma and alpha planes,
387 and of @code{min(cw,ch)/2} for the chroma planes.
389 @var{luma_power}, @var{chroma_power}, and @var{alpha_power} represent
390 how many times the boxblur filter is applied to the corresponding
393 Some examples follow:
398 Apply a boxblur filter with luma, chroma, and alpha radius
405 Set luma radius to 2, alpha and chroma radius to 0
411 Set luma and chroma radius to a fraction of the video dimension
413 boxblur=min(h\,w)/10:1:min(cw\,ch)/10:1
420 Copy the input source unchanged to the output. Mainly useful for
425 Crop the input video to @var{out_w}:@var{out_h}:@var{x}:@var{y}.
427 The parameters are expressions containing the following constants:
431 the corresponding mathematical approximated values for e
432 (euler number), pi (greek PI), PHI (golden ratio)
435 the computed values for @var{x} and @var{y}. They are evaluated for
439 the input width and heigth
442 same as @var{in_w} and @var{in_h}
445 the output (cropped) width and heigth
448 same as @var{out_w} and @var{out_h}
451 same as @var{iw} / @var{ih}
454 input sample aspect ratio
457 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
460 horizontal and vertical chroma subsample values. For example for the
461 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
464 the number of input frame, starting from 0
467 the position in the file of the input frame, NAN if unknown
470 timestamp expressed in seconds, NAN if the input timestamp is unknown
474 The @var{out_w} and @var{out_h} parameters specify the expressions for
475 the width and height of the output (cropped) video. They are
476 evaluated just at the configuration of the filter.
478 The default value of @var{out_w} is "in_w", and the default value of
479 @var{out_h} is "in_h".
481 The expression for @var{out_w} may depend on the value of @var{out_h},
482 and the expression for @var{out_h} may depend on @var{out_w}, but they
483 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
484 evaluated after @var{out_w} and @var{out_h}.
486 The @var{x} and @var{y} parameters specify the expressions for the
487 position of the top-left corner of the output (non-cropped) area. They
488 are evaluated for each frame. If the evaluated value is not valid, it
489 is approximated to the nearest valid value.
491 The default value of @var{x} is "(in_w-out_w)/2", and the default
492 value for @var{y} is "(in_h-out_h)/2", which set the cropped area at
493 the center of the input image.
495 The expression for @var{x} may depend on @var{y}, and the expression
496 for @var{y} may depend on @var{x}.
498 Follow some examples:
500 # crop the central input area with size 100x100
503 # crop the central input area with size 2/3 of the input video
504 "crop=2/3*in_w:2/3*in_h"
506 # crop the input video central square
509 # delimit the rectangle with the top-left corner placed at position
510 # 100:100 and the right-bottom corner corresponding to the right-bottom
511 # corner of the input image.
512 crop=in_w-100:in_h-100:100:100
514 # crop 10 pixels from the left and right borders, and 20 pixels from
515 # the top and bottom borders
516 "crop=in_w-2*10:in_h-2*20"
518 # keep only the bottom right quarter of the input image
519 "crop=in_w/2:in_h/2:in_w/2:in_h/2"
521 # crop height for getting Greek harmony
522 "crop=in_w:1/PHI*in_w"
525 "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)"
527 # erratic camera effect depending on timestamp
528 "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)"
530 # set x depending on the value of y
531 "crop=in_w/2:in_h/2:y:10+10*sin(n/10)"
536 Auto-detect crop size.
538 Calculate necessary cropping parameters and prints the recommended
539 parameters through the logging system. The detected dimensions
540 correspond to the non-black area of the input video.
542 It accepts the syntax:
544 cropdetect[=@var{limit}[:@var{round}[:@var{reset}]]]
550 Threshold, which can be optionally specified from nothing (0) to
551 everything (255), defaults to 24.
554 Value which the width/height should be divisible by, defaults to
555 16. The offset is automatically adjusted to center the video. Use 2 to
556 get only even dimensions (needed for 4:2:2 video). 16 is best when
557 encoding to most video codecs.
560 Counter that determines after how many frames cropdetect will reset
561 the previously detected largest video area and start over to detect
562 the current optimal crop area. Defaults to 0.
564 This can be useful when channel logos distort the video area. 0
565 indicates never reset and return the largest area encountered during
571 Suppress a TV station logo by a simple interpolation of the surrounding
572 pixels. Just set a rectangle covering the logo and watch it disappear
573 (and sometimes something even uglier appear - your mileage may vary).
575 The filter accepts parameters as a string of the form
576 "@var{x}:@var{y}:@var{w}:@var{h}:@var{band}", or as a list of
577 @var{key}=@var{value} pairs, separated by ":".
579 The description of the accepted parameters follows.
584 Specify the top left corner coordinates of the logo. They must be
588 Specify the width and height of the logo to clear. They must be
592 Specify the thickness of the fuzzy edge of the rectangle (added to
593 @var{w} and @var{h}). The default value is 4.
596 When set to 1, a green rectangle is drawn on the screen to simplify
597 finding the right @var{x}, @var{y}, @var{w}, @var{h} parameters, and
598 @var{band} is set to 4. The default value is 0.
602 Some examples follow.
607 Set a rectangle covering the area with top left corner coordinates 0,0
608 and size 100x77, setting a band of size 10:
614 As the previous example, but use named options:
616 delogo=x=0:y=0:w=100:h=77:band=10
623 Draw a colored box on the input image.
625 It accepts the syntax:
627 drawbox=@var{x}:@var{y}:@var{width}:@var{height}:@var{color}
633 Specify the top left corner coordinates of the box. Default to 0.
636 Specify the width and height of the box, if 0 they are interpreted as
637 the input width and height. Default to 0.
640 Specify the color of the box to write, it can be the name of a color
641 (case insensitive match) or a 0xRRGGBB[AA] sequence.
644 Follow some examples:
646 # draw a black box around the edge of the input image
649 # draw a box with color red and an opacity of 50%
650 drawbox=10:20:200:60:red@@0.5"
655 Draw text string or text from specified file on top of video using the
658 To enable compilation of this filter you need to configure FFmpeg with
659 @code{--enable-libfreetype}.
661 The filter also recognizes strftime() sequences in the provided text
662 and expands them accordingly. Check the documentation of strftime().
664 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
667 The description of the accepted parameters follows.
672 The font file to be used for drawing text. Path must be included.
673 This parameter is mandatory.
676 The text string to be drawn. The text must be a sequence of UTF-8
678 This parameter is mandatory if no file is specified with the parameter
682 A text file containing text to be drawn. The text must be a sequence
683 of UTF-8 encoded characters.
685 This parameter is mandatory if no text string is specified with the
686 parameter @var{text}.
688 If both text and textfile are specified, an error is thrown.
691 The offsets where text will be drawn within the video frame.
692 Relative to the top/left border of the output image.
694 The default value of @var{x} and @var{y} is 0.
697 The font size to be used for drawing text.
698 The default value of @var{fontsize} is 16.
701 The color to be used for drawing fonts.
702 Either a string (e.g. "red") or in 0xRRGGBB[AA] format
703 (e.g. "0xff000033"), possibly followed by an alpha specifier.
704 The default value of @var{fontcolor} is "black".
707 The color to be used for drawing box around text.
708 Either a string (e.g. "yellow") or in 0xRRGGBB[AA] format
709 (e.g. "0xff00ff"), possibly followed by an alpha specifier.
710 The default value of @var{boxcolor} is "white".
713 Used to draw a box around text using background color.
714 Value should be either 1 (enable) or 0 (disable).
715 The default value of @var{box} is 0.
717 @item shadowx, shadowy
718 The x and y offsets for the text shadow position with respect to the
719 position of the text. They can be either positive or negative
720 values. Default value for both is "0".
723 The color to be used for drawing a shadow behind the drawn text. It
724 can be a color name (e.g. "yellow") or a string in the 0xRRGGBB[AA]
725 form (e.g. "0xff00ff"), possibly followed by an alpha specifier.
726 The default value of @var{shadowcolor} is "black".
729 Flags to be used for loading the fonts.
731 The flags map the corresponding flags supported by libfreetype, and are
732 a combination of the following values:
739 @item vertical_layout
743 @item ignore_global_advance_width
745 @item ignore_transform
752 Default value is "render".
754 For more information consult the documentation for the FT_LOAD_*
758 The size in number of spaces to use for rendering the tab.
762 For example the command:
764 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
767 will draw "Test Text" with font FreeSerif, using the default values
768 for the optional parameters.
772 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
773 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
776 will draw 'Test Text' with font FreeSerif of size 24 at position x=100
777 and y=50 (counting from the top-left corner of the screen), text is
778 yellow with a red box around it. Both the text and the box have an
781 Note that the double quotes are not necessary if spaces are not used
782 within the parameter list.
784 For more information about libfreetype, check:
785 @url{http://www.freetype.org/}.
789 Apply fade-in/out effect to input video.
791 It accepts the parameters:
792 @var{type}:@var{start_frame}:@var{nb_frames}
794 @var{type} specifies if the effect type, can be either "in" for
795 fade-in, or "out" for a fade-out effect.
797 @var{start_frame} specifies the number of the start frame for starting
798 to apply the fade effect.
800 @var{nb_frames} specifies the number of frames for which the fade
801 effect has to last. At the end of the fade-in effect the output video
802 will have the same intensity as the input video, at the end of the
803 fade-out transition the output video will be completely black.
805 A few usage examples follow, usable too as test scenarios.
807 # fade in first 30 frames of video
810 # fade out last 45 frames of a 200-frame video
813 # fade in first 25 frames and fade out last 25 frames of a 1000-frame video
814 fade=in:0:25, fade=out:975:25
816 # make first 5 frames black, then fade in from frame 5-24
822 Transform the field order of the input video.
824 It accepts one parameter which specifies the required field order that
825 the input interlaced video will be transformed to. The parameter can
826 assume one of the following values:
830 output bottom field first
832 output top field first
835 Default value is "tff".
837 Transformation is achieved by shifting the picture content up or down
838 by one line, and filling the remaining line with appropriate picture content.
839 This method is consistent with most broadcast field order converters.
841 If the input video is not flagged as being interlaced, or it is already
842 flagged as being of the required output field order then this filter does
843 not alter the incoming video.
845 This filter is very useful when converting to or from PAL DV material,
846 which is bottom field first.
850 ./ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
855 Buffer input images and send them when they are requested.
857 This filter is mainly useful when auto-inserted by the libavfilter
860 The filter does not take parameters.
864 Convert the input video to one of the specified pixel formats.
865 Libavfilter will try to pick one that is supported for the input to
868 The filter accepts a list of pixel format names, separated by ":",
869 for example "yuv420p:monow:rgb24".
871 Some examples follow:
873 # convert the input video to the format "yuv420p"
876 # convert the input video to any of the formats in the list
877 format=yuv420p:yuv444p:yuv410p
883 Apply a frei0r effect to the input video.
885 To enable compilation of this filter you need to install the frei0r
886 header and configure FFmpeg with --enable-frei0r.
888 The filter supports the syntax:
890 @var{filter_name}[@{:|=@}@var{param1}:@var{param2}:...:@var{paramN}]
893 @var{filter_name} is the name to the frei0r effect to load. If the
894 environment variable @env{FREI0R_PATH} is defined, the frei0r effect
895 is searched in each one of the directories specified by the colon
896 separated list in @env{FREIOR_PATH}, otherwise in the standard frei0r
897 paths, which are in this order: @file{HOME/.frei0r-1/lib/},
898 @file{/usr/local/lib/frei0r-1/}, @file{/usr/lib/frei0r-1/}.
900 @var{param1}, @var{param2}, ... , @var{paramN} specify the parameters
901 for the frei0r effect.
903 A frei0r effect parameter can be a boolean (whose values are specified
904 with "y" and "n"), a double, a color (specified by the syntax
905 @var{R}/@var{G}/@var{B}, @var{R}, @var{G}, and @var{B} being float
906 numbers from 0.0 to 1.0) or by an @code{av_parse_color()} color
907 description), a position (specified by the syntax @var{X}/@var{Y},
908 @var{X} and @var{Y} being float numbers) and a string.
910 The number and kind of parameters depend on the loaded effect. If an
911 effect parameter is not specified the default value is set.
913 Some examples follow:
915 # apply the distort0r effect, set the first two double parameters
916 frei0r=distort0r:0.5:0.01
918 # apply the colordistance effect, takes a color as first parameter
919 frei0r=colordistance:0.2/0.3/0.4
920 frei0r=colordistance:violet
921 frei0r=colordistance:0x112233
923 # apply the perspective effect, specify the top left and top right
925 frei0r=perspective:0.2/0.2:0.8/0.2
928 For more information see:
929 @url{http://piksel.org/frei0r}
933 Fix the banding artifacts that are sometimes introduced into nearly flat
934 regions by truncation to 8bit colordepth.
935 Interpolate the gradients that should go where the bands are, and
938 This filter is designed for playback only. Do not use it prior to
939 lossy compression, because compression tends to lose the dither and
940 bring back the bands.
942 The filter takes two optional parameters, separated by ':':
943 @var{strength}:@var{radius}
945 @var{strength} is the maximum amount by which the filter will change
946 any one pixel. Also the threshold for detecting nearly flat
947 regions. Acceptable values range from .51 to 255, default value is
948 1.2, out-of-range values will be clipped to the valid range.
950 @var{radius} is the neighborhood to fit the gradient to. A larger
951 radius makes for smoother gradients, but also prevents the filter from
952 modifying the pixels near detailed regions. Acceptable values are
953 8-32, default value is 16, out-of-range values will be clipped to the
966 Flip the input video horizontally.
968 For example to horizontally flip the video in input with
971 ffmpeg -i in.avi -vf "hflip" out.avi
976 High precision/quality 3d denoise filter. This filter aims to reduce
977 image noise producing smooth images and making still images really
978 still. It should enhance compressibility.
980 It accepts the following optional parameters:
981 @var{luma_spatial}:@var{chroma_spatial}:@var{luma_tmp}:@var{chroma_tmp}
985 a non-negative float number which specifies spatial luma strength,
989 a non-negative float number which specifies spatial chroma strength,
990 defaults to 3.0*@var{luma_spatial}/4.0
993 a float number which specifies luma temporal strength, defaults to
994 6.0*@var{luma_spatial}/4.0
997 a float number which specifies chroma temporal strength, defaults to
998 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
1001 @section lut, lutrgb, lutyuv
1003 Compute a look-up table for binding each pixel component input value
1004 to an output value, and apply it to input video.
1006 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
1007 to an RGB input video.
1009 These filters accept in input a ":"-separated list of options, which
1010 specify the expressions used for computing the lookup table for the
1011 corresponding pixel component values.
1013 The @var{lut} filter requires either YUV or RGB pixel formats in
1014 input, and accepts the options:
1016 @var{c0} (first pixel component)
1017 @var{c1} (second pixel component)
1018 @var{c2} (third pixel component)
1019 @var{c3} (fourth pixel component, corresponds to the alpha component)
1022 The exact component associated to each option depends on the format in
1025 The @var{lutrgb} filter requires RGB pixel formats in input, and
1026 accepts the options:
1028 @var{r} (red component)
1029 @var{g} (green component)
1030 @var{b} (blue component)
1031 @var{a} (alpha component)
1034 The @var{lutyuv} filter requires YUV pixel formats in input, and
1035 accepts the options:
1037 @var{y} (Y/luminance component)
1038 @var{u} (U/Cb component)
1039 @var{v} (V/Cr component)
1040 @var{a} (alpha component)
1043 The expressions can contain the following constants and functions:
1047 the corresponding mathematical approximated values for e
1048 (euler number), pi (greek PI), PHI (golden ratio)
1051 the input width and heigth
1054 input value for the pixel component
1057 the input value clipped in the @var{minval}-@var{maxval} range
1060 maximum value for the pixel component
1063 minimum value for the pixel component
1066 the negated value for the pixel component value clipped in the
1067 @var{minval}-@var{maxval} range , it corresponds to the expression
1068 "maxval-clipval+minval"
1071 the computed value in @var{val} clipped in the
1072 @var{minval}-@var{maxval} range
1074 @item gammaval(gamma)
1075 the computed gamma correction value of the pixel component value
1076 clipped in the @var{minval}-@var{maxval} range, corresponds to the
1078 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
1082 All expressions default to "val".
1084 Some examples follow:
1086 # negate input video
1087 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
1088 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
1090 # the above is the same as
1091 lutrgb="r=negval:g=negval:b=negval"
1092 lutyuv="y=negval:u=negval:v=negval"
1097 # remove chroma components, turns the video into a graytone image
1098 lutyuv="u=128:v=128"
1100 # apply a luma burning effect
1103 # remove green and blue components
1106 # set a constant alpha channel value on input
1107 format=rgba,lutrgb=a="maxval-minval/2"
1109 # correct luminance gamma by a 0.5 factor
1110 lutyuv=y=gammaval(0.5)
1115 Apply an MPlayer filter to the input video.
1117 This filter provides a wrapper around most of the filters of
1120 This wrapper is considered experimental. Some of the wrapped filters
1121 may not work properly and we may drop support for them, as they will
1122 be implemented natively into FFmpeg. Thus you should avoid
1123 depending on them when writing portable scripts.
1125 The filters accepts the parameters:
1126 @var{filter_name}[:=]@var{filter_params}
1128 @var{filter_name} is the name of a supported MPlayer filter,
1129 @var{filter_params} is a string containing the parameters accepted by
1132 The list of the currently supported filters follows:
1186 The parameter syntax and behavior for the listed filters are the same
1187 of the corresponding MPlayer filters. For detailed instructions check
1188 the "VIDEO FILTERS" section in the MPlayer manual.
1190 Some examples follow:
1192 # remove a logo by interpolating the surrounding pixels
1193 mp=delogo=200:200:80:20:1
1195 # adjust gamma, brightness, contrast
1198 # tweak hue and saturation
1202 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
1208 This filter accepts an integer in input, if non-zero it negates the
1209 alpha component (if available). The default value in input is 0.
1213 Force libavfilter not to use any of the specified pixel formats for the
1214 input to the next filter.
1216 The filter accepts a list of pixel format names, separated by ":",
1217 for example "yuv420p:monow:rgb24".
1219 Some examples follow:
1221 # force libavfilter to use a format different from "yuv420p" for the
1222 # input to the vflip filter
1223 noformat=yuv420p,vflip
1225 # convert the input video to any of the formats not contained in the list
1226 noformat=yuv420p:yuv444p:yuv410p
1231 Pass the video source unchanged to the output.
1235 Apply video transform using libopencv.
1237 To enable this filter install libopencv library and headers and
1238 configure FFmpeg with --enable-libopencv.
1240 The filter takes the parameters: @var{filter_name}@{:=@}@var{filter_params}.
1242 @var{filter_name} is the name of the libopencv filter to apply.
1244 @var{filter_params} specifies the parameters to pass to the libopencv
1245 filter. If not specified the default values are assumed.
1247 Refer to the official libopencv documentation for more precise
1249 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
1251 Follows the list of supported libopencv filters.
1256 Dilate an image by using a specific structuring element.
1257 This filter corresponds to the libopencv function @code{cvDilate}.
1259 It accepts the parameters: @var{struct_el}:@var{nb_iterations}.
1261 @var{struct_el} represents a structuring element, and has the syntax:
1262 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
1264 @var{cols} and @var{rows} represent the number of colums and rows of
1265 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
1266 point, and @var{shape} the shape for the structuring element, and
1267 can be one of the values "rect", "cross", "ellipse", "custom".
1269 If the value for @var{shape} is "custom", it must be followed by a
1270 string of the form "=@var{filename}". The file with name
1271 @var{filename} is assumed to represent a binary image, with each
1272 printable character corresponding to a bright pixel. When a custom
1273 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
1274 or columns and rows of the read file are assumed instead.
1276 The default value for @var{struct_el} is "3x3+0x0/rect".
1278 @var{nb_iterations} specifies the number of times the transform is
1279 applied to the image, and defaults to 1.
1281 Follow some example:
1283 # use the default values
1286 # dilate using a structuring element with a 5x5 cross, iterate two times
1287 ocv=dilate=5x5+2x2/cross:2
1289 # read the shape from the file diamond.shape, iterate two times
1290 # the file diamond.shape may contain a pattern of characters like this:
1296 # the specified cols and rows are ignored (but not the anchor point coordinates)
1297 ocv=0x0+2x2/custom=diamond.shape:2
1302 Erode an image by using a specific structuring element.
1303 This filter corresponds to the libopencv function @code{cvErode}.
1305 The filter accepts the parameters: @var{struct_el}:@var{nb_iterations},
1306 with the same syntax and semantics as the @ref{dilate} filter.
1310 Smooth the input video.
1312 The filter takes the following parameters:
1313 @var{type}:@var{param1}:@var{param2}:@var{param3}:@var{param4}.
1315 @var{type} is the type of smooth filter to apply, and can be one of
1316 the following values: "blur", "blur_no_scale", "median", "gaussian",
1317 "bilateral". The default value is "gaussian".
1319 @var{param1}, @var{param2}, @var{param3}, and @var{param4} are
1320 parameters whose meanings depend on smooth type. @var{param1} and
1321 @var{param2} accept integer positive values or 0, @var{param3} and
1322 @var{param4} accept float values.
1324 The default value for @var{param1} is 3, the default value for the
1325 other parameters is 0.
1327 These parameters correspond to the parameters assigned to the
1328 libopencv function @code{cvSmooth}.
1332 Overlay one video on top of another.
1334 It takes two inputs and one output, the first input is the "main"
1335 video on which the second input is overlayed.
1337 It accepts the parameters: @var{x}:@var{y}.
1339 @var{x} is the x coordinate of the overlayed video on the main video,
1340 @var{y} is the y coordinate. The parameters are expressions containing
1341 the following parameters:
1344 @item main_w, main_h
1345 main input width and height
1348 same as @var{main_w} and @var{main_h}
1350 @item overlay_w, overlay_h
1351 overlay input width and height
1354 same as @var{overlay_w} and @var{overlay_h}
1357 Be aware that frames are taken from each input video in timestamp
1358 order, hence, if their initial timestamps differ, it is a a good idea
1359 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
1360 have them begin in the same zero timestamp, as it does the example for
1361 the @var{movie} filter.
1363 Follow some examples:
1365 # draw the overlay at 10 pixels from the bottom right
1366 # corner of the main video.
1367 overlay=main_w-overlay_w-10:main_h-overlay_h-10
1369 # insert a transparent PNG logo in the bottom left corner of the input
1370 movie=logo.png [logo];
1371 [in][logo] overlay=10:main_h-overlay_h-10 [out]
1373 # insert 2 different transparent PNG logos (second logo on bottom
1375 movie=logo1.png [logo1];
1376 movie=logo2.png [logo2];
1377 [in][logo1] overlay=10:H-h-10 [in+logo1];
1378 [in+logo1][logo2] overlay=W-w-10:H-h-10 [out]
1380 # add a transparent color layer on top of the main video,
1381 # WxH specifies the size of the main input to the overlay filter
1382 color=red@.3:WxH [over]; [in][over] overlay [out]
1385 You can chain togheter more overlays but the efficiency of such
1386 approach is yet to be tested.
1390 Add paddings to the input image, and places the original input at the
1391 given coordinates @var{x}, @var{y}.
1393 It accepts the following parameters:
1394 @var{width}:@var{height}:@var{x}:@var{y}:@var{color}.
1396 The parameters @var{width}, @var{height}, @var{x}, and @var{y} are
1397 expressions containing the following constants:
1401 the corresponding mathematical approximated values for e
1402 (euler number), pi (greek PI), phi (golden ratio)
1405 the input video width and heigth
1408 same as @var{in_w} and @var{in_h}
1411 the output width and heigth, that is the size of the padded area as
1412 specified by the @var{width} and @var{height} expressions
1415 same as @var{out_w} and @var{out_h}
1418 x and y offsets as specified by the @var{x} and @var{y}
1419 expressions, or NAN if not yet specified
1422 same as @var{iw} / @var{ih}
1425 input sample aspect ratio
1428 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
1431 horizontal and vertical chroma subsample values. For example for the
1432 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1435 Follows the description of the accepted parameters.
1440 Specify the size of the output image with the paddings added. If the
1441 value for @var{width} or @var{height} is 0, the corresponding input size
1442 is used for the output.
1444 The @var{width} expression can reference the value set by the
1445 @var{height} expression, and viceversa.
1447 The default value of @var{width} and @var{height} is 0.
1451 Specify the offsets where to place the input image in the padded area
1452 with respect to the top/left border of the output image.
1454 The @var{x} expression can reference the value set by the @var{y}
1455 expression, and viceversa.
1457 The default value of @var{x} and @var{y} is 0.
1461 Specify the color of the padded area, it can be the name of a color
1462 (case insensitive match) or a 0xRRGGBB[AA] sequence.
1464 The default value of @var{color} is "black".
1468 Some examples follow:
1471 # Add paddings with color "violet" to the input video. Output video
1472 # size is 640x480, the top-left corner of the input video is placed at
1474 pad=640:480:0:40:violet
1476 # pad the input to get an output with dimensions increased bt 3/2,
1477 # and put the input video at the center of the padded area
1478 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
1480 # pad the input to get a squared output with size equal to the maximum
1481 # value between the input width and height, and put the input video at
1482 # the center of the padded area
1483 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
1485 # pad the input to get a final w/h ratio of 16:9
1486 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
1488 # for anamorphic video, in order to set the output display aspect ratio,
1489 # it is necessary to use sar in the expression, according to the relation:
1490 # (ih * X / ih) * sar = output_dar
1491 # X = output_dar / sar
1492 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
1494 # double output size and put the input video in the bottom-right
1495 # corner of the output padded area
1496 pad="2*iw:2*ih:ow-iw:oh-ih"
1499 @section pixdesctest
1501 Pixel format descriptor test filter, mainly useful for internal
1502 testing. The output video should be equal to the input video.
1506 format=monow, pixdesctest
1509 can be used to test the monowhite pixel format descriptor definition.
1513 Scale the input video to @var{width}:@var{height} and/or convert the image format.
1515 The parameters @var{width} and @var{height} are expressions containing
1516 the following constants:
1520 the corresponding mathematical approximated values for e
1521 (euler number), pi (greek PI), phi (golden ratio)
1524 the input width and heigth
1527 same as @var{in_w} and @var{in_h}
1530 the output (cropped) width and heigth
1533 same as @var{out_w} and @var{out_h}
1536 same as @var{iw} / @var{ih}
1539 input sample aspect ratio
1542 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
1545 horizontal and vertical chroma subsample values. For example for the
1546 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1549 If the input image format is different from the format requested by
1550 the next filter, the scale filter will convert the input to the
1553 If the value for @var{width} or @var{height} is 0, the respective input
1554 size is used for the output.
1556 If the value for @var{width} or @var{height} is -1, the scale filter will
1557 use, for the respective output size, a value that maintains the aspect
1558 ratio of the input image.
1560 The default value of @var{width} and @var{height} is 0.
1562 Some examples follow:
1564 # scale the input video to a size of 200x100.
1567 # scale the input to 2x
1569 # the above is the same as
1572 # scale the input to half size
1575 # increase the width, and set the height to the same size
1578 # seek for Greek harmony
1582 # increase the height, and set the width to 3/2 of the height
1585 # increase the size, but make the size a multiple of the chroma
1586 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
1588 # increase the width to a maximum of 500 pixels, keep the same input aspect ratio
1589 scale='min(500\, iw*3/2):-1'
1593 Select frames to pass in output.
1595 It accepts in input an expression, which is evaluated for each input
1596 frame. If the expression is evaluated to a non-zero value, the frame
1597 is selected and passed to the output, otherwise it is discarded.
1599 The expression can contain the following constants:
1612 the sequential number of the filtered frame, starting from 0
1615 the sequential number of the selected frame, starting from 0
1617 @item prev_selected_n
1618 the sequential number of the last selected frame, NAN if undefined
1621 timebase of the input timestamps
1624 the PTS (Presentation TimeStamp) of the filtered video frame,
1625 expressed in @var{TB} units, NAN if undefined
1628 the PTS (Presentation TimeStamp) of the filtered video frame,
1629 expressed in seconds, NAN if undefined
1632 the PTS of the previously filtered video frame, NAN if undefined
1634 @item prev_selected_pts
1635 the PTS of the last previously filtered video frame, NAN if undefined
1637 @item prev_selected_t
1638 the PTS of the last previously selected video frame, NAN if undefined
1641 the PTS of the first video frame in the video, NAN if undefined
1644 the time of the first video frame in the video, NAN if undefined
1647 the picture type of the filtered frame, can assume one of the following
1659 @item interlace_type
1660 the frame interlace type, can assume one of the following values:
1662 @item INTERLACE_TYPE_P
1663 the frame is progressive (not interlaced)
1664 @item INTERLACE_TYPE_T
1665 the frame is top-field-first
1666 @item INTERLACE_TYPE_B
1667 the frame is bottom-field-first
1671 1 if the filtered frame is a key-frame, 0 otherwise
1674 the position in the file of the filtered frame, -1 if the information
1675 is not available (e.g. for synthetic video)
1678 The default value of the select expression is "1".
1680 Some examples follow:
1683 # select all frames in input
1686 # the above is the same as:
1692 # select only I-frames
1693 select='eq(pict_type\,PICT_TYPE_I)'
1695 # select one frame every 100
1696 select='not(mod(n\,100))'
1698 # select only frames contained in the 10-20 time interval
1699 select='gte(t\,10)*lte(t\,20)'
1701 # select only I frames contained in the 10-20 time interval
1702 select='gte(t\,10)*lte(t\,20)*eq(pict_type\,PICT_TYPE_I)'
1704 # select frames with a minimum distance of 10 seconds
1705 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
1711 Set the Display Aspect Ratio for the filter output video.
1713 This is done by changing the specified Sample (aka Pixel) Aspect
1714 Ratio, according to the following equation:
1715 @math{DAR = HORIZONTAL_RESOLUTION / VERTICAL_RESOLUTION * SAR}
1717 Keep in mind that this filter does not modify the pixel dimensions of
1718 the video frame. Also the display aspect ratio set by this filter may
1719 be changed by later filters in the filterchain, e.g. in case of
1720 scaling or if another "setdar" or a "setsar" filter is applied.
1722 The filter accepts a parameter string which represents the wanted
1723 display aspect ratio.
1724 The parameter can be a floating point number string, or an expression
1725 of the form @var{num}:@var{den}, where @var{num} and @var{den} are the
1726 numerator and denominator of the aspect ratio.
1727 If the parameter is not specified, it is assumed the value "0:1".
1729 For example to change the display aspect ratio to 16:9, specify:
1732 # the above is equivalent to
1736 See also the @ref{setsar} filter documentation.
1740 Change the PTS (presentation timestamp) of the input video frames.
1742 Accept in input an expression evaluated through the eval API, which
1743 can contain the following constants:
1747 the presentation timestamp in input
1759 the count of the input frame, starting from 0.
1762 the PTS of the first video frame
1765 tell if the current frame is interlaced
1768 original position in the file of the frame, or undefined if undefined
1769 for the current frame
1779 Some examples follow:
1782 # start counting PTS from zero
1794 # fixed rate 25 fps with some jitter
1795 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
1801 Set the Sample (aka Pixel) Aspect Ratio for the filter output video.
1803 Note that as a consequence of the application of this filter, the
1804 output display aspect ratio will change according to the following
1806 @math{DAR = HORIZONTAL_RESOLUTION / VERTICAL_RESOLUTION * SAR}
1808 Keep in mind that the sample aspect ratio set by this filter may be
1809 changed by later filters in the filterchain, e.g. if another "setsar"
1810 or a "setdar" filter is applied.
1812 The filter accepts a parameter string which represents the wanted
1813 sample aspect ratio.
1814 The parameter can be a floating point number string, or an expression
1815 of the form @var{num}:@var{den}, where @var{num} and @var{den} are the
1816 numerator and denominator of the aspect ratio.
1817 If the parameter is not specified, it is assumed the value "0:1".
1819 For example to change the sample aspect ratio to 10:11, specify:
1826 Set the timebase to use for the output frames timestamps.
1827 It is mainly useful for testing timebase configuration.
1829 It accepts in input an arithmetic expression representing a rational.
1830 The expression can contain the constants "PI", "E", "PHI", "AVTB" (the
1831 default timebase), and "intb" (the input timebase).
1833 The default value for the input is "intb".
1835 Follow some examples.
1838 # set the timebase to 1/25
1841 # set the timebase to 1/10
1844 #set the timebase to 1001/1000
1847 #set the timebase to 2*intb
1850 #set the default timebase value
1856 Show a line containing various information for each input video frame.
1857 The input video is not modified.
1859 The shown line contains a sequence of key/value pairs of the form
1860 @var{key}:@var{value}.
1862 A description of each shown parameter follows:
1866 sequential number of the input frame, starting from 0
1869 Presentation TimeStamp of the input frame, expressed as a number of
1870 time base units. The time base unit depends on the filter input pad.
1873 Presentation TimeStamp of the input frame, expressed as a number of
1877 position of the frame in the input stream, -1 if this information in
1878 unavailable and/or meanigless (for example in case of synthetic video)
1884 sample aspect ratio of the input frame, expressed in the form
1888 size of the input frame, expressed in the form
1889 @var{width}x@var{height}
1892 interlaced mode ("P" for "progressive", "T" for top field first, "B"
1893 for bottom field first)
1896 1 if the frame is a key frame, 0 otherwise
1899 picture type of the input frame ("I" for an I-frame, "P" for a
1900 P-frame, "B" for a B-frame, "?" for unknown type).
1901 Check also the documentation of the @code{AVPictureType} enum and of
1902 the @code{av_get_picture_type_char} function defined in
1903 @file{libavutil/avutil.h}.
1906 Adler-32 checksum of all the planes of the input frame
1908 @item plane_checksum
1909 Adler-32 checksum of each plane of the input frame, expressed in the form
1910 "[@var{c0} @var{c1} @var{c2} @var{c3}]"
1915 Pass the images of input video on to next video filter as multiple
1919 ./ffmpeg -i in.avi -vf "slicify=32" out.avi
1922 The filter accepts the slice height as parameter. If the parameter is
1923 not specified it will use the default value of 16.
1925 Adding this in the beginning of filter chains should make filtering
1926 faster due to better use of the memory cache.
1930 Pass on the input video to two outputs. Both outputs are identical to
1935 [in] split [splitout1][splitout2];
1936 [splitout1] crop=100:100:0:0 [cropout];
1937 [splitout2] pad=200:200:100:100 [padout];
1940 will create two separate outputs from the same input, one cropped and
1945 Transpose rows with columns in the input video and optionally flip it.
1947 It accepts a parameter representing an integer, which can assume the
1952 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
1960 Rotate by 90 degrees clockwise, that is:
1968 Rotate by 90 degrees counterclockwise, that is:
1976 Rotate by 90 degrees clockwise and vertically flip, that is:
1986 Sharpen or blur the input video.
1988 It accepts the following parameters:
1989 @var{luma_msize_x}:@var{luma_msize_y}:@var{luma_amount}:@var{chroma_msize_x}:@var{chroma_msize_y}:@var{chroma_amount}
1991 Negative values for the amount will blur the input video, while positive
1992 values will sharpen. All parameters are optional and default to the
1993 equivalent of the string '5:5:1.0:5:5:0.0'.
1998 Set the luma matrix horizontal size. It can be an integer between 3
1999 and 13, default value is 5.
2002 Set the luma matrix vertical size. It can be an integer between 3
2003 and 13, default value is 5.
2006 Set the luma effect strength. It can be a float number between -2.0
2007 and 5.0, default value is 1.0.
2009 @item chroma_msize_x
2010 Set the chroma matrix horizontal size. It can be an integer between 3
2011 and 13, default value is 5.
2013 @item chroma_msize_y
2014 Set the chroma matrix vertical size. It can be an integer between 3
2015 and 13, default value is 5.
2018 Set the chroma effect strength. It can be a float number between -2.0
2019 and 5.0, default value is 0.0.
2024 # Strong luma sharpen effect parameters
2027 # Strong blur of both luma and chroma parameters
2028 unsharp=7:7:-2:7:7:-2
2030 # Use the default values with @command{ffmpeg}
2031 ./ffmpeg -i in.avi -vf "unsharp" out.mp4
2036 Flip the input video vertically.
2039 ./ffmpeg -i in.avi -vf "vflip" out.avi
2044 Deinterlace the input video ("yadif" means "yet another deinterlacing
2047 It accepts the optional parameters: @var{mode}:@var{parity}:@var{auto}.
2049 @var{mode} specifies the interlacing mode to adopt, accepts one of the
2054 output 1 frame for each frame
2056 output 1 frame for each field
2058 like 0 but skips spatial interlacing check
2060 like 1 but skips spatial interlacing check
2065 @var{parity} specifies the picture field parity assumed for the input
2066 interlaced video, accepts one of the following values:
2070 assume top field first
2072 assume bottom field first
2074 enable automatic detection
2077 Default value is -1.
2078 If interlacing is unknown or decoder does not export this information,
2079 top field first will be assumed.
2081 @var{auto} specifies if deinterlacer should trust the interlaced flag
2082 and only deinterlace frames marked as interlaced
2086 deinterlace all frames
2088 only deinterlace frames marked as interlaced
2093 @c man end VIDEO FILTERS
2095 @chapter Video Sources
2096 @c man begin VIDEO SOURCES
2098 Below is a description of the currently available video sources.
2102 Buffer video frames, and make them available to the filter chain.
2104 This source is mainly intended for a programmatic use, in particular
2105 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
2107 It accepts the following parameters:
2108 @var{width}:@var{height}:@var{pix_fmt_string}:@var{timebase_num}:@var{timebase_den}:@var{sample_aspect_ratio_num}:@var{sample_aspect_ratio.den}:@var{scale_params}
2110 All the parameters but @var{scale_params} need to be explicitely
2113 Follows the list of the accepted parameters.
2118 Specify the width and height of the buffered video frames.
2120 @item pix_fmt_string
2121 A string representing the pixel format of the buffered video frames.
2122 It may be a number corresponding to a pixel format, or a pixel format
2125 @item timebase_num, timebase_den
2126 Specify numerator and denomitor of the timebase assumed by the
2127 timestamps of the buffered frames.
2129 @item sample_aspect_ratio.num, sample_aspect_ratio.den
2130 Specify numerator and denominator of the sample aspect ratio assumed
2131 by the video frames.
2134 Specify the optional parameters to be used for the scale filter which
2135 is automatically inserted when an input change is detected in the
2136 input size or format.
2141 buffer=320:240:yuv410p:1:24:1:1
2144 will instruct the source to accept video frames with size 320x240 and
2145 with format "yuv410p", assuming 1/24 as the timestamps timebase and
2146 square pixels (1:1 sample aspect ratio).
2147 Since the pixel format with name "yuv410p" corresponds to the number 6
2148 (check the enum PixelFormat definition in @file{libavutil/pixfmt.h}),
2149 this example corresponds to:
2151 buffer=320:240:6:1:24:1:1
2156 Provide an uniformly colored input.
2158 It accepts the following parameters:
2159 @var{color}:@var{frame_size}:@var{frame_rate}
2161 Follows the description of the accepted parameters.
2166 Specify the color of the source. It can be the name of a color (case
2167 insensitive match) or a 0xRRGGBB[AA] sequence, possibly followed by an
2168 alpha specifier. The default value is "black".
2171 Specify the size of the sourced video, it may be a string of the form
2172 @var{width}x@var{heigth}, or the name of a size abbreviation. The
2173 default value is "320x240".
2176 Specify the frame rate of the sourced video, as the number of frames
2177 generated per second. It has to be a string in the format
2178 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
2179 number or a valid video frame rate abbreviation. The default value is
2184 For example the following graph description will generate a red source
2185 with an opacity of 0.2, with size "qcif" and a frame rate of 10
2186 frames per second, which will be overlayed over the source connected
2187 to the pad with identifier "in".
2190 "color=red@@0.2:qcif:10 [color]; [in][color] overlay [out]"
2195 Read a video stream from a movie container.
2197 It accepts the syntax: @var{movie_name}[:@var{options}] where
2198 @var{movie_name} is the name of the resource to read (not necessarily
2199 a file but also a device or a stream accessed through some protocol),
2200 and @var{options} is an optional sequence of @var{key}=@var{value}
2201 pairs, separated by ":".
2203 The description of the accepted options follows.
2207 @item format_name, f
2208 Specifies the format assumed for the movie to read, and can be either
2209 the name of a container or an input device. If not specified the
2210 format is guessed from @var{movie_name} or by probing.
2212 @item seek_point, sp
2213 Specifies the seek point in seconds, the frames will be output
2214 starting from this seek point, the parameter is evaluated with
2215 @code{av_strtod} so the numerical value may be suffixed by an IS
2216 postfix. Default value is "0".
2218 @item stream_index, si
2219 Specifies the index of the video stream to read. If the value is -1,
2220 the best suited video stream will be automatically selected. Default
2225 This filter allows to overlay a second video on top of main input of
2226 a filtergraph as shown in this graph:
2228 input -----------> deltapts0 --> overlay --> output
2231 movie --> scale--> deltapts1 -------+
2234 Some examples follow:
2236 # skip 3.2 seconds from the start of the avi file in.avi, and overlay it
2237 # on top of the input labelled as "in".
2238 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [movie];
2239 [in] setpts=PTS-STARTPTS, [movie] overlay=16:16 [out]
2241 # read from a video4linux2 device, and overlay it on top of the input
2243 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [movie];
2244 [in] setpts=PTS-STARTPTS, [movie] overlay=16:16 [out]
2250 Generate various test patterns, as generated by the MPlayer test filter.
2252 The size of the generated video is fixed, and is 256x256.
2253 This source is useful in particular for testing encoding features.
2255 This source accepts an optional sequence of @var{key}=@var{value} pairs,
2256 separated by ":". The description of the accepted options follows.
2261 Specify the frame rate of the sourced video, as the number of frames
2262 generated per second. It has to be a string in the format
2263 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
2264 number or a valid video frame rate abbreviation. The default value is
2268 Set the video duration of the sourced video. The accepted syntax is:
2270 [-]HH[:MM[:SS[.m...]]]
2273 See also the function @code{av_parse_time()}.
2275 If not specified, or the expressed duration is negative, the video is
2276 supposed to be generated forever.
2280 Set the number or the name of the test to perform. Supported tests are:
2295 Default value is "all", which will cycle through the list of all tests.
2298 For example the following:
2303 will generate a "dc_luma" test pattern.
2307 Null video source, never return images. It is mainly useful as a
2308 template and to be employed in analysis / debugging tools.
2310 It accepts as optional parameter a string of the form
2311 @var{width}:@var{height}:@var{timebase}.
2313 @var{width} and @var{height} specify the size of the configured
2314 source. The default values of @var{width} and @var{height} are
2315 respectively 352 and 288 (corresponding to the CIF size format).
2317 @var{timebase} specifies an arithmetic expression representing a
2318 timebase. The expression can contain the constants "PI", "E", "PHI",
2319 "AVTB" (the default timebase), and defaults to the value "AVTB".
2323 Provide a frei0r source.
2325 To enable compilation of this filter you need to install the frei0r
2326 header and configure FFmpeg with --enable-frei0r.
2328 The source supports the syntax:
2330 @var{size}:@var{rate}:@var{src_name}[@{=|:@}@var{param1}:@var{param2}:...:@var{paramN}]
2333 @var{size} is the size of the video to generate, may be a string of the
2334 form @var{width}x@var{height} or a frame size abbreviation.
2335 @var{rate} is the rate of the video to generate, may be a string of
2336 the form @var{num}/@var{den} or a frame rate abbreviation.
2337 @var{src_name} is the name to the frei0r source to load. For more
2338 information regarding frei0r and how to set the parameters read the
2339 section @ref{frei0r} in the description of the video filters.
2341 Some examples follow:
2343 # generate a frei0r partik0l source with size 200x200 and framerate 10
2344 # which is overlayed on the overlay filter main input
2345 frei0r_src=200x200:10:partik0l=1234 [overlay]; [in][overlay] overlay
2348 @section rgbtestsrc, testsrc
2350 The @code{rgbtestsrc} source generates an RGB test pattern useful for
2351 detecting RGB vs BGR issues. You should see a red, green and blue
2352 stripe from top to bottom.
2354 The @code{testsrc} source generates a test video pattern, showing a
2355 color pattern, a scrolling gradient and a timestamp. This is mainly
2356 intended for testing purposes.
2358 Both sources accept an optional sequence of @var{key}=@var{value} pairs,
2359 separated by ":". The description of the accepted options follows.
2364 Specify the size of the sourced video, it may be a string of the form
2365 @var{width}x@var{heigth}, or the name of a size abbreviation. The
2366 default value is "320x240".
2369 Specify the frame rate of the sourced video, as the number of frames
2370 generated per second. It has to be a string in the format
2371 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
2372 number or a valid video frame rate abbreviation. The default value is
2376 Set the sample aspect ratio of the sourced video.
2379 Set the video duration of the sourced video. The accepted syntax is:
2381 [-]HH[:MM[:SS[.m...]]]
2384 See also the function @code{av_parse_time()}.
2386 If not specified, or the expressed duration is negative, the video is
2387 supposed to be generated forever.
2390 For example the following:
2392 testsrc=duration=5.3:size=qcif:rate=10
2395 will generate a video with a duration of 5.3 seconds, with size
2396 176x144 and a framerate of 10 frames per second.
2398 @c man end VIDEO SOURCES
2400 @chapter Video Sinks
2401 @c man begin VIDEO SINKS
2403 Below is a description of the currently available video sinks.
2407 Buffer video frames, and make them available to the end of the filter
2410 This sink is mainly intended for a programmatic use, in particular
2411 through the interface defined in @file{libavfilter/buffersink.h}.
2413 It does not require a string parameter in input, but you need to
2414 specify a pointer to a list of supported pixel formats terminated by
2415 -1 in the opaque parameter provided to @code{avfilter_init_filter}
2416 when initializing this sink.
2420 Null video sink, do absolutely nothing with the input video. It is
2421 mainly useful as a template and to be employed in analysis / debugging
2424 @c man end VIDEO SINKS