ffmpeg.1

arm平台下的H264编码和解码源代码

.RS 4
.IP "\fB0\fR" 4
.IX Item "0"
\&\s-1FF_IDCT_AUTO\s0 (default)
.IP "\fB1\fR" 4
.IX Item "1"
\&\s-1FF_IDCT_INT\s0          
.IP "\fB2\fR" 4
.IX Item "2"
\&\s-1FF_IDCT_SIMPLE\s0       
.IP "\fB3\fR" 4
.IX Item "3"
\&\s-1FF_IDCT_SIMPLEMMX\s0    
.IP "\fB4\fR" 4
.IX Item "4"
\&\s-1FF_IDCT_LIBMPEG2MMX\s0  
.IP "\fB5\fR" 4
.IX Item "5"
\&\s-1FF_IDCT_PS2\s0          
.IP "\fB6\fR" 4
.IX Item "6"
\&\s-1FF_IDCT_MLIB\s0         
.IP "\fB7\fR" 4
.IX Item "7"
\&\s-1FF_IDCT_ARM\s0          
.IP "\fB8\fR" 4
.IX Item "8"
\&\s-1FF_IDCT_ALTIVEC\s0      
.IP "\fB9\fR" 4
.IX Item "9"
\&\s-1FF_IDCT_SH4\s0          
.IP "\fB10\fR" 4
.IX Item "10"
\&\s-1FF_IDCT_SIMPLEARM\s0    
.RE
.RS 4
.RE
.IP "\fB\-er n\fR" 4
.IX Item "-er n"
set error resilience to \fIn\fR.
.RS 4
.IP "\fB1\fR" 4
.IX Item "1"
\&\s-1FF_ER_CAREFULL\s0 (default)
.IP "\fB2\fR" 4
.IX Item "2"
\&\s-1FF_ER_COMPLIANT\s0
.IP "\fB3\fR" 4
.IX Item "3"
\&\s-1FF_ER_AGGRESSIVE\s0
.IP "\fB4\fR" 4
.IX Item "4"
\&\s-1FF_ER_VERY_AGGRESSIVE\s0
.RE
.RS 4
.RE
.IP "\fB\-ec bit_mask\fR" 4
.IX Item "-ec bit_mask"
set error concealment to \fIbit_mask\fR. \fIbit_mask\fR is a bit mask of
the following values:
.RS 4
.IP "\fB1\fR" 4
.IX Item "1"
\&\s-1FF_EC_GUESS_MVS\s0 (default=enabled)
.IP "\fB2\fR" 4
.IX Item "2"
\&\s-1FF_EC_DEBLOCK\s0 (default=enabled)
.RE
.RS 4
.RE
.IP "\fB\-bf frames\fR" 4
.IX Item "-bf frames"
use 'frames' B frames (supported for \s-1MPEG\-1\s0, \s-1MPEG\-2\s0 and \s-1MPEG\-4\s0)
.IP "\fB\-mbd mode\fR" 4
.IX Item "-mbd mode"
macroblock decision
.RS 4
.IP "\fB0\fR" 4
.IX Item "0"
\&\s-1FF_MB_DECISION_SIMPLE:\s0 use mb_cmp (cannot change it yet in ffmpeg)
.IP "\fB1\fR" 4
.IX Item "1"
\&\s-1FF_MB_DECISION_BITS:\s0 chooses the one which needs the fewest bits
.IP "\fB2\fR" 4
.IX Item "2"
\&\s-1FF_MB_DECISION_RD:\s0 rate distoration
.RE
.RS 4
.RE
.IP "\fB\-4mv\fR" 4
.IX Item "-4mv"
use four motion vector by macroblock (only \s-1MPEG\-4\s0)
.IP "\fB\-part\fR" 4
.IX Item "-part"
use data partitioning (only \s-1MPEG\-4\s0)
.IP "\fB\-bug param\fR" 4
.IX Item "-bug param"
workaround not auto detected encoder bugs
.IP "\fB\-strict strictness\fR" 4
.IX Item "-strict strictness"
how strictly to follow the standarts
.IP "\fB\-aic\fR" 4
.IX Item "-aic"
enable Advanced intra coding (h263+)
.IP "\fB\-umv\fR" 4
.IX Item "-umv"
enable Unlimited Motion Vector (h263+)
.IP "\fB\-deinterlace\fR" 4
.IX Item "-deinterlace"
deinterlace pictures
.IP "\fB\-interlace\fR" 4
.IX Item "-interlace"
force interlacing support in encoder (only \s-1MPEG\-2\s0 and \s-1MPEG\-4\s0). Use this option
if your input file is interlaced and if you want to keep the interlaced
format for minimum losses. The alternative is to deinterlace the input
stream with \fB\-deinterlace\fR, but deinterlacing introduces more
losses.
.IP "\fB\-psnr\fR" 4
.IX Item "-psnr"
calculate \s-1PSNR\s0 of compressed frames
.IP "\fB\-vstats\fR" 4
.IX Item "-vstats"
dump video coding statistics to \fIvstats_HHMMSS.log\fR.
.IP "\fB\-vhook module\fR" 4
.IX Item "-vhook module"
insert video processing \fImodule\fR. \fImodule\fR contains the module
name and its parameters separated by spaces.
.Sh "Audio Options"
.IX Subsection "Audio Options"
.IP "\fB\-ab bitrate\fR" 4
.IX Item "-ab bitrate"
set audio bitrate (in kbit/s)
.IP "\fB\-ar freq\fR" 4
.IX Item "-ar freq"
set the audio sampling freq (default = 44100 Hz)
.IP "\fB\-ab bitrate\fR" 4
.IX Item "-ab bitrate"
set the audio bitrate in kbit/s (default = 64)
.IP "\fB\-ac channels\fR" 4
.IX Item "-ac channels"
set the number of audio channels (default = 1)
.IP "\fB\-an\fR" 4
.IX Item "-an"
disable audio recording
.IP "\fB\-acodec codec\fR" 4
.IX Item "-acodec codec"
force audio codec to \fIcodec\fR. Use the \f(CW\*(C`copy\*(C'\fR special value to
tell that the raw codec data must be copied as is.
.Sh "Audio/Video grab options"
.IX Subsection "Audio/Video grab options"
.IP "\fB\-vd device\fR" 4
.IX Item "-vd device"
set video grab device (e.g. \fI/dev/video0\fR)
.IP "\fB\-vc channel\fR" 4
.IX Item "-vc channel"
set video grab channel (\s-1DV1394\s0 only)
.IP "\fB\-tvstd standard\fR" 4
.IX Item "-tvstd standard"
set television standard (\s-1NTSC\s0, \s-1PAL\s0 (\s-1SECAM\s0))
.IP "\fB\-dv1394\fR" 4
.IX Item "-dv1394"
set \s-1DV1394\s0 grab
.IP "\fB\-ad device\fR" 4
.IX Item "-ad device"
set audio device (e.g. \fI/dev/dsp\fR)
.Sh "Advanced options"
.IX Subsection "Advanced options"
.IP "\fB\-map file:stream\fR" 4
.IX Item "-map file:stream"
set input stream mapping
.IP "\fB\-debug\fR" 4
.IX Item "-debug"
print specific debug info
.IP "\fB\-benchmark\fR" 4
.IX Item "-benchmark"
add timings for benchmarking
.IP "\fB\-hex\fR" 4
.IX Item "-hex"
dump each input packet
.IP "\fB\-bitexact\fR" 4
.IX Item "-bitexact"
only use bit exact algorithms (for codec testing)
.IP "\fB\-ps size\fR" 4
.IX Item "-ps size"
set packet size in bits
.IP "\fB\-re\fR" 4
.IX Item "-re"
read input at native frame rate. Mainly used to simulate a grab device.
.IP "\fB\-loop\fR" 4
.IX Item "-loop"
loop over the input stream. Currently it works only for image
streams. This option is used for ffserver automatic testing.
.Sh "FFmpeg formula evaluator"
.IX Subsection "FFmpeg formula evaluator"
When evaluating a rate control string, FFmpeg uses an internal formula
evaluator. 
.PP
The following binary operators are available: \f(CW\*(C`+\*(C'\fR, \f(CW\*(C`\-\*(C'\fR,
\&\f(CW\*(C`*\*(C'\fR, \f(CW\*(C`/\*(C'\fR, \f(CW\*(C`^\*(C'\fR.
.PP
The following unary operators are available: \f(CW\*(C`+\*(C'\fR, \f(CW\*(C`\-\*(C'\fR,
\&\f(CW\*(C`(...)\*(C'\fR.
.PP
The following functions are available:
.IP "\fIsinh(x)\fR" 4
.IX Item "sinh(x)"
.PD 0
.IP "\fIcosh(x)\fR" 4
.IX Item "cosh(x)"
.IP "\fItanh(x)\fR" 4
.IX Item "tanh(x)"
.IP "\fIsin(x)\fR" 4
.IX Item "sin(x)"
.IP "\fIcos(x)\fR" 4
.IX Item "cos(x)"
.IP "\fItan(x)\fR" 4
.IX Item "tan(x)"
.IP "\fIexp(x)\fR" 4
.IX Item "exp(x)"
.IP "\fIlog(x)\fR" 4
.IX Item "log(x)"
.IP "\fIsquish(x)\fR" 4
.IX Item "squish(x)"
.IP "\fIgauss(x)\fR" 4
.IX Item "gauss(x)"
.IP "\fIabs(x)\fR" 4
.IX Item "abs(x)"
.IP "\fImax(x, y)\fR" 4
.IX Item "max(x, y)"
.IP "\fImin(x, y)\fR" 4
.IX Item "min(x, y)"
.IP "\fIgt(x, y)\fR" 4
.IX Item "gt(x, y)"
.IP "\fIlt(x, y)\fR" 4
.IX Item "lt(x, y)"
.IP "\fIeq(x, y)\fR" 4
.IX Item "eq(x, y)"
.IP "\fIbits2qp(bits)\fR" 4
.IX Item "bits2qp(bits)"
.IP "\fIqp2bits(qp)\fR" 4
.IX Item "qp2bits(qp)"
.PD
.PP
The following constants are available:
.IP "\fI\s-1PI\s0\fR" 4
.IX Item "PI"
.PD 0
.IP "\fIE\fR" 4
.IX Item "E"
.IP "\fIiTex\fR" 4
.IX Item "iTex"
.IP "\fIpTex\fR" 4
.IX Item "pTex"
.IP "\fItex\fR" 4
.IX Item "tex"
.IP "\fImv\fR" 4
.IX Item "mv"
.IP "\fIfCode\fR" 4
.IX Item "fCode"
.IP "\fIiCount\fR" 4
.IX Item "iCount"
.IP "\fImcVar\fR" 4
.IX Item "mcVar"
.IP "\fIvar\fR" 4
.IX Item "var"
.IP "\fIisI\fR" 4
.IX Item "isI"
.IP "\fIisP\fR" 4
.IX Item "isP"
.IP "\fIisB\fR" 4
.IX Item "isB"
.IP "\fIavgQP\fR" 4
.IX Item "avgQP"
.IP "\fIqComp\fR" 4
.IX Item "qComp"
.IP "\fIavgIITex\fR" 4
.IX Item "avgIITex"
.IP "\fIavgPITex\fR" 4
.IX Item "avgPITex"
.IP "\fIavgPPTex\fR" 4
.IX Item "avgPPTex"
.IP "\fIavgBPTex\fR" 4
.IX Item "avgBPTex"
.IP "\fIavgTex\fR" 4
.IX Item "avgTex"
.PD
.SH "EXAMPLES"
.IX Header "EXAMPLES"
.Sh "Video and Audio grabbing"
.IX Subsection "Video and Audio grabbing"
FFmpeg can use a video4linux compatible video source and any Open Sound
System audio source:
.PP
.Vb 1
\&        ffmpeg /tmp/out.mpg
.Ve
.PP
Note that you must activate the right video source and channel before
launching ffmpeg. You can use any \s-1TV\s0 viewer such as xawtv
(<\fBhttp://bytesex.org/xawtv/\fR>) by Gerd Knorr which I find very
good. You must also set correctly the audio recording levels with a
standard mixer.
.Sh "Video and Audio file format conversion"
.IX Subsection "Video and Audio file format conversion"
* ffmpeg can use any supported file format and protocol as input: 
.PP
Examples:
.PP
* You can input from \s-1YUV\s0 files:
.PP
.Vb 1
\&        ffmpeg -i /tmp/test%d.Y /tmp/out.mpg
.Ve
.PP
It will use the files: 
.PP
.Vb 2
\&        /tmp/test0.Y, /tmp/test0.U, /tmp/test0.V,
\&        /tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc...
.Ve
.PP
The Y files use twice the resolution of the U and V files. They are
raw files, without header. They can be generated by all decent video
decoders. You must specify the size of the image with the \fB\-s\fR option
if ffmpeg cannot guess it.
.PP
* You can input from a \s-1RAW\s0 \s-1YUV420P\s0 file:
.PP
.Vb 1
\&        ffmpeg -i /tmp/test.yuv /tmp/out.avi
.Ve
.PP
The \s-1RAW\s0 \s-1YUV420P\s0 is a file containing \s-1RAW\s0 \s-1YUV\s0 planar, for each frame first
come the Y plane followed by U and V planes, which are half vertical and
horizontal resolution.
.PP
* You can output to a \s-1RAW\s0 \s-1YUV420P\s0 file:
.PP
.Vb 1
\&        ffmpeg -i mydivx.avi -o hugefile.yuv
.Ve
.PP
* You can set several input files and output files:
.PP
.Vb 1
\&        ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg
.Ve
.PP
Convert the audio file a.wav and the raw yuv video file a.yuv
to mpeg file a.mpg
.PP
* You can also do audio and video conversions at the same time:
.PP
.Vb 1
\&        ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2
.Ve
.PP
Convert the sample rate of a.wav to 22050 Hz and encode it to \s-1MPEG\s0 audio.
.PP
* You can encode to several formats at the same time and define a
mapping from input stream to output streams:
.PP
.Vb 1
\&        ffmpeg -i /tmp/a.wav -ab 64 /tmp/a.mp2 -ab 128 /tmp/b.mp2 -map 0:0 -map 0:0
.Ve
.PP
Convert a.wav to a.mp2 at 64 kbits and b.mp2 at 128 kbits. '\-map
file:index' specify which input stream is used for each output
stream, in the order of the definition of output streams.
.PP
* You can transcode decrypted VOBs
.PP
.Vb 1
\&        ffmpeg -i snatch_1.vob -f avi -vcodec mpeg4 -b 800 -g 300 -bf 2 -acodec mp3 -ab 128 snatch.avi
.Ve
.PP
This is a typical \s-1DVD\s0 ripper example, input from a \s-1VOB\s0 file, output
to an \s-1AVI\s0 file with \s-1MPEG\-4\s0 video and \s-1MP3\s0 audio, note that in this
command we use B frames so the \s-1MPEG\-4\s0 stream is DivX5 compatible, \s-1GOP\s0
size is 300 that means an \s-1INTRA\s0 frame every 10 seconds for 29.97 fps
input video.  Also the audio stream is \s-1MP3\s0 encoded so you need \s-1LAME\s0
support which is enabled using \f(CW\*(C`\-\-enable\-mp3lame\*(C'\fR when
configuring.  The mapping is particularly useful for \s-1DVD\s0 transcoding
to get the desired audio language.
.PP
\&\s-1NOTE:\s0 to see the supported input formats, use \f(CW\*(C`ffmpeg \-formats\*(C'\fR.
.SH "SEE ALSO"
.IX Header "SEE ALSO"
\&\fIffserver\fR\|(1), \fIffplay\fR\|(1) and the html documentation of \fIffmpeg\fR.
.SH "AUTHOR"
.IX Header "AUTHOR"
Fabrice Bellard