Recherche avancée

Sur d’autres sites (11252)

• Normalize video contrast to the full dynamic range with ffmpeg ?

  18 mai 2017, par Roofus

  The following ImageMagick command line normalizes the RGB channels of an image individually, so that in each channel, the smallest value maps to 0 and the largest value maps to 255 :

  convert fish.jpg -channel all -contrast-stretch 0.003x0.003% fish2.jpg

  fish.jpg
  fish2.jpg

  Is there an ffmpeg filter which can normalize the RGB channels of every individual frame of a video ?

  The only filter I can find is "histeq", which equalizes (flattens) rather than normalizes the contrast. I have applied it per RGB channel using variations on this command line :

  ffmpeg -i fish.jpg -vf "format=rgb24,extractplanes=r+g+b[r][g][b],[r]histeq=strength=.1[r2],[g]histeq=strength=.1[g2],[b]histeq=strength=.1[b2],[g2][b2][r2]mergeplanes=0x001020:gbrp" fish3.jpg

  but since it tries to flatten the histogram, it always gives a much different (unacceptable) result ; for example, the imgur image whose code is eHL51.jpg (not enough reputation for link).

  Based on this answer :
  video normalization with ffmpeg
  I have also tested -vf "scale=out_range=full" and -vf "pp=al:f" (see the log below) but in both cases the result fish3.jpg was unchanged from fish.jpg, so apparently nothing was done.

  C:\Users\Roofus\Desktop>ffmpeg -i fish.jpg -vf scale=out_range=full -color_range 2 -pix_fmt yuvj420p fish3.jpg
  
  ffmpeg version 3.2.4 Copyright (c) 2000-2017 the FFmpeg developers
  
    built with gcc 6.3.0 (GCC)
  
    configuration: --enable-gpl --enable-version3 --enable-d3d11va --enable-dxva2 --enable-libmfx --enable-nvenc --enable-
  
  avisynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enable-iconv --enable-libass --enable-libb
  
  luray --enable-libbs2b --enable-libcaca --enable-libfreetype --enable-libgme --enable-libgsm --enable-libilbc --enable-l
  
  ibmodplug --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenh264 --enable-libope
  
  njpeg --enable-libopus --enable-librtmp --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libtheora --enabl
  
  e-libtwolame --enable-libvidstab --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx --enable-libwavpack --enable
  
  -libwebp --enable-libx264 --enable-libx265 --enable-libxavs --enable-libxvid --enable-libzimg --enable-lzma --enable-zli
  
  b
  
    libavutil      55. 34.101 / 55. 34.101
  
    libavcodec     57. 64.101 / 57. 64.101
  
    libavformat    57. 56.101 / 57. 56.101
  
    libavdevice    57.  1.100 / 57.  1.100
  
    libavfilter     6. 65.100 /  6. 65.100
  
    libswscale      4.  2.100 /  4.  2.100
  
    libswresample   2.  3.100 /  2.  3.100
  
    libpostproc    54.  1.100 / 54.  1.100
  
  Input #0, image2, from 'fish.jpg':
  
    Duration: 00:00:00.04, start: 0.000000, bitrate: 3942 kb/s
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc, bt470bg/unknown/unknown), 960x540 [SAR 96:96 DAR 16:9], 25 tbr, 25 tbn, 25 t
  
  bc
  
  [swscaler @ 0000000000442f60] deprecated pixel format used, make sure you did set range correctly
  
  Output #0, image2, to 'fish3.jpg':
  
    Metadata:
  
      encoder         : Lavf57.56.101
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc), 960x540 [SAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc
  
      Metadata:
  
        encoder         : Lavc57.64.101 mjpeg
  
      Side data:
  
        cpb: bitrate max/min/avg: 0/0/200000 buffer size: 0 vbv_delay: -1
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (mjpeg (native) -> mjpeg (native))
  
  Press [q] to stop, [?] for help
  
  frame=    1 fps=0.0 q=2.2 Lsize=N/A time=00:00:00.04 bitrate=N/A speed=   5x
  
  video:23kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
  
  
  
  C:\Users\Roofus\Desktop>fish3.jpg
  
  
  
  C:\Users\Roofus\Desktop>ffmpeg -i fish.jpg -vf "pp=al:f" -color_range 2 -pix_fmt yuvj420p fish3.jpg
  
  ffmpeg version 3.2.4 Copyright (c) 2000-2017 the FFmpeg developers
  
    built with gcc 6.3.0 (GCC)
  
    configuration: --enable-gpl --enable-version3 --enable-d3d11va --enable-dxva2 --enable-libmfx --enable-nvenc --enable-
  
  avisynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enable-iconv --enable-libass --enable-libb
  
  luray --enable-libbs2b --enable-libcaca --enable-libfreetype --enable-libgme --enable-libgsm --enable-libilbc --enable-l
  
  ibmodplug --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenh264 --enable-libope
  
  njpeg --enable-libopus --enable-librtmp --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libtheora --enabl
  
  e-libtwolame --enable-libvidstab --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx --enable-libwavpack --enable
  
  -libwebp --enable-libx264 --enable-libx265 --enable-libxavs --enable-libxvid --enable-libzimg --enable-lzma --enable-zli
  
  b
  
    libavutil      55. 34.101 / 55. 34.101
  
    libavcodec     57. 64.101 / 57. 64.101
  
    libavformat    57. 56.101 / 57. 56.101
  
    libavdevice    57.  1.100 / 57.  1.100
  
    libavfilter     6. 65.100 /  6. 65.100
  
    libswscale      4.  2.100 /  4.  2.100
  
    libswresample   2.  3.100 /  2.  3.100
  
    libpostproc    54.  1.100 / 54.  1.100
  
  Input #0, image2, from 'fish.jpg':
  
    Duration: 00:00:00.04, start: 0.000000, bitrate: 3942 kb/s
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc, bt470bg/unknown/unknown), 960x540 [SAR 96:96 DAR 16:9], 25 tbr, 25 tbn, 25 t
  
  bc
  
  File 'fish3.jpg' already exists. Overwrite ? [y/N] y
  
  [Parsed_pp_0 @ 0000000000321da0] This syntax is deprecated. Use '|' to separate the list items.
  
  Output #0, image2, to 'fish3.jpg':
  
    Metadata:
  
      encoder         : Lavf57.56.101
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc), 960x540 [SAR 96:96 DAR 16:9], q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc
  
      Metadata:
  
        encoder         : Lavc57.64.101 mjpeg
  
      Side data:
  
        cpb: bitrate max/min/avg: 0/0/200000 buffer size: 0 vbv_delay: -1
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (mjpeg (native) -> mjpeg (native))
  
  Press [q] to stop, [?] for help
  
  frame=    1 fps=0.0 q=2.2 Lsize=N/A time=00:00:00.04 bitrate=N/A speed=4.44x
  
  video:23kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
  
  
  
  C:\Users\Roofus\Desktop>fish3.jpg

• lsws/utils : Make gray10 and gray12 full-scale like gray8 and gray16.

  4 avril 2017, par Carl Eugen Hoyos

  lsws/utils : Make gray10 and gray12 full-scale like gray8 and gray16.
  

  • [DH] libswscale/utils.c
  • [DH] libswscale/version.h

• FFmpeg : real-time buffer[ ] [input] too full or near too full (101% of size : 3041280 [rtbufsize parameter]) frame dropped

  5 janvier 2017, par lostin2010

  I use the library of ffmpeg to decode stream from [TTQ HD Camera] and encode it to a rtmp stream.
  but I receive a lot of warnings like the picture below.
  i try to set qmin and qmax , it seems a little better. but still not totally resolve the problem.

  encoder_context->qmin = 10;
  
  encoder_context->qmax = 51;

  who knows this is why ?

  [dshow @ 04bfc640] real-time buffer [TTQ HD Camera] [video input] too full or near too full (101% of size: 3041280 [rtbufsize parameter])! frame dropped!