Recherche avancée

Sur d’autres sites (8336)

• ffmpeg command to copy raw data into MP4 hangs and never finishes

  28 mai 2019, par James Adams

  I am running an ffmpeg command to copy raw H.265 data into an MP4 file and the command never completes. The file isn’t large, just 10 seconds worth of data. Here’s the command I’m running :

  $ ffmpeg -rtsp_transport tcp -i rtsp://user:pass@71.185.124.195:554/c1/b1558830329/e1558830339/replay/ -vcodec copy -y test_clip.mp4

  I then get output like this :

  ffmpeg version 3.4.6-0ubuntu0.18.04.1 Copyright (c) 2000-2019 the FFmpeg developers
  
    built with gcc 7 (Ubuntu 7.3.0-16ubuntu3)
  
    configuration: --prefix=/usr --extra-version=0ubuntu0.18.04.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --disable-stripping --enable-avresample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librubberband --enable-librsvg --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzmq --enable-libzvbi --enable-omx --enable-openal --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-chromaprint --enable-frei0r --enable-libopencv --enable-libx264 --enable-shared
  
    libavutil      55. 78.100 / 55. 78.100
  
    libavcodec     57.107.100 / 57.107.100
  
    libavformat    57. 83.100 / 57. 83.100
  
    libavdevice    57. 10.100 / 57. 10.100
  
    libavfilter     6.107.100 /  6.107.100
  
    libavresample   3.  7.  0 /  3.  7.  0
  
    libswscale      4.  8.100 /  4.  8.100
  
    libswresample   2.  9.100 /  2.  9.100
  
    libpostproc    54.  7.100 / 54.  7.100
  
  Guessed Channel Layout for Input Stream #0.1 : mono
  
  Input #0, rtsp, from 'rtsp://user:pass@71.85.104.195:554/c1/b1558830329/e1558830339/replay/':
  
    Metadata:
  
      title           : ONVIF RTSP Server
  
    Duration: N/A, start: 0.000000, bitrate: N/A
  
      Stream #0:0: Video: h264 (High), yuvj420p(pc, bt709, progressive), 1920x1080, 30 fps, 30 tbr, 90k tbn, 60 tbc
  
      Stream #0:1: Audio: pcm_mulaw, 8000 Hz, mono, s16, 64 kb/s
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (copy)
  
    Stream #0:1 -> #0:1 (pcm_mulaw (native) -> aac (native))
  
  Press [q] to stop, [?] for help
  
  [aac @ 0x55b71ce31900] Too many bits 8832.000000 > 6144 per frame requested, clamping to max
  
  Output #0, mp4, to 'test_clip.mp4':
  
    Metadata:
  
      title           : ONVIF RTSP Server
  
      encoder         : Lavf57.83.100
  
      Stream #0:0: Video: h264 (High) (avc1 / 0x31637661), yuvj420p(pc, bt709, progressive), 1920x1080, q=2-31, 30 fps, 30 tbr, 90k tbn, 90k tbc
  
      Stream #0:1: Audio: aac (LC) (mp4a / 0x6134706D), 8000 Hz, mono, fltp, 48 kb/s
  
      Metadata:
  
        encoder         : Lavc57.107.100 aac
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18000, current: 3000; changing to 18001. This may result in incorrect timestamps in the output file.
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18001, current: 6000; changing to 18002. This may result in incorrect timestamps in the output file.
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18002, current: 9000; changing to 18003. This may result in incorrect timestamps in the output file.
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18003, current: 12000; changing to 18004. This may result in incorrect timestamps in the output file.
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18004, current: 15000; changing to 18005. This may result in incorrect timestamps in the output file.
  
  [mp4 @ 0x55b71ce17e00] Non-monotonous DTS in output stream 0:0; previous: 18005, current: 18000; changing to 18006. This may result in incorrect timestamps in the output file.
  
  frame=   44 fps=0.0 q=-1.0 size=     256kB time=00:00:01.43 bitrate=1463.4kbits/frame=   60 fps= 57 q=-1.0 size=     512kB time=00:00:01.96 bitrate=2132.9kbits/frame=   76 fps= 48 q=-1.0 size=     768kB time=00:00:02.50 bitrate=2516.7kbits/frame=   92 fps= 44 q=-1.0 size=    1024kB time=00:00:03.03 bitrate=2765.6kbits/frame=  108 fps= 41 q=-1.0 size=    1024kB time=00:00:03.56 bitrate=2352.0kbits/[NULL @ 0x55b71cdfa540] SEI type 5 size 2208 truncated at 1944
  
  frame=  123 fps= 39 q=-1.0 size=    1280kB time=00:00:04.06 bitrate=2578.6kbits/frame=  139 fps= 38 q=-1.0 size=    1536kB time=00:00:04.60 bitrate=2735.5kbits/frame=  155 fps= 37 q=-1.0 size=    1536kB time=00:00:05.13 bitrate=2451.3kbits/frame=  171 fps= 36 q=-1.0 size=    1792kB time=00:00:05.66 bitrate=2590.7kbits/frame=  187 fps= 36 q=-1.0 size=    2048kB time=00:00:06.20 bitrate=2706.1kbits/frame=  203 fps= 35 q=-1.0 size=    2304kB time=00:00:06.73 bitrate=2803.2kbits/frame=  219 fps= 35 q=-1.0 size=    2304kB time=00:00:07.26 bitrate=2597.4kbits/frame=  235 fps= 34 q=-1.0 size=    2560kB time=00:00:07.80 bitrate=2688.7kbits/frame=  246 fps= 33 q=-1.0 size=    2560kB time=00:00:08.16 bitrate=2568.0kbits/frame=  267 fps= 34 q=-1.0 size=    3072kB time=00:00:08.86 bitrate=2838.3kbits/frame=  282 fps= 34 q=-1.0 size=    3072kB time=00:00:09.36 bitrate=2686.8kbits/frame=  298 fps= 33 q=-1.0 size=    3328kB time=00:00:09.90 bitrate=2753.9kbits/frame=  314 fps= 33 q=-1.0 size=    3328kB time=00:00:10.43 bitrate=2613.1kbits/^Cspeed=1.11x    

  The command never completes and I need to kill it using Ctrl-C.

  I have also tried adding the options -nostdin -loglevel error and appending this to the end of the command : > /dev/null 2>&1 < /dev/null but to no avail.

  I am testing the above since it mimics the actual code I’m developing which utilizes a Python package that wraps calls to ffmpeg. The Python code below works well on a laptop but hangs on EC2 (both are Ubuntu 18.04) :

  import argparse
  
  import datetime
  
  import ffmpeg
  
  
  
  
  
  # ------------------------------------------------------------------------------
  
  if __name__ == "__main__":
  
  
  
      # USAGE
  
      # $ python collect_and_store.py --rtsp rtsp://user:pass1@71.85.125.110:554 \
  
      #       --duration 30 --count 10
  
  
  
      # construct the argument parser and parse the arguments
  
      args_parser = argparse.ArgumentParser()
  
      args_parser.add_argument("--rtsp",
  
                               required=True,
  
                               type=str,
  
                               help="RTSP URL for video stream")
  
      args_parser.add_argument("--duration",
  
                               required=True,
  
                               type=int,
  
                               help="duration of saved clips (in seconds)")
  
      args_parser.add_argument("--count",
  
                               required=True,
  
                               type=int,
  
                               help="number of clips to save")
  
      args = vars(args_parser.parse_args())
  
  
  
      # sanity check for some of the arguments
  
      if not args["rtsp"].lower().startswith("rtsp://"):
  
          raise ValueError("Invalid input URL -- only RTSP supported")
  
  
  
      seconds_per_clip = args["duration"]
  
      start = int(datetime.datetime.now().strftime("%s"))
  
      end = start + seconds_per_clip
  
      number_of_files_to_collect = args["count"]
  
  
  
      while number_of_files_to_collect > 0:
  
  
  
          # build URL with start and end times
  
          # NOTE URL is for Uniview RTSP, add options for other camera types
  
          url = args["rtsp"] + f"/c1/b{start}/e{end}/replay/"
  
  
  
          # file where we'll write clip data
  
          temp_file = f"clip_b{start}_e{end}.mp4"
  
  
  
          # create the equivalent of the ffmpeg command:
  
          # $ ffmpeg -i  -vcodec copy -y -rtsp_transport tcp 
  
          stream = ffmpeg.input(url)
  
          stream = ffmpeg.output(stream, temp_file,
  
                                 **{"codec:v": "copy",
  
                                    "rtsp_transport": "tcp",
  
                                    "y": None
  
                                    }
  
                                 )
  
          ffmpeg.run(stream)
  
  
  
          print(f"\n\nMP4 file created: {temp_file}")
  
  
  
          number_of_files_to_collect -= 1
  
          start = end + 1
  
          end = start + seconds_per_clip

• swscale : Add support for NV24 and NV42

  10 mai 2019, par Philip Langdale

  swscale : Add support for NV24 and NV42
  The implementation is pretty straight-forward. Most of the existing
  
  NV12 codepaths work regardless of subsampling and are re-used as is.
  
  Where necessary I wrote the slightly different NV24 versions.
  Finally, the one thing that confused me for a long time was the
  
  asm specific x86 path that did an explicit exclusion check for NV12.
  
  I replaced that with a semi-planar check and also updated the
  
  equivalent PPC code, which Lauri kindly checked.
  

  • [DH] libswscale/input.c
  • [DH] libswscale/output.c
  • [DH] libswscale/ppc/swscale_altivec.c
  • [DH] libswscale/ppc/swscale_vsx.c
  • [DH] libswscale/swscale_unscaled.c
  • [DH] libswscale/utils.c
  • [DH] libswscale/version.h
  • [DH] libswscale/x86/swscale_template.c
  • [DH] tests/ref/fate/filter-pixfmts-copy
  • [DH] tests/ref/fate/filter-pixfmts-crop
  • [DH] tests/ref/fate/filter-pixfmts-field
  • [DH] tests/ref/fate/filter-pixfmts-fieldorder
  • [DH] tests/ref/fate/filter-pixfmts-hflip
  • [DH] tests/ref/fate/filter-pixfmts-il
  • [DH] tests/ref/fate/filter-pixfmts-null
  • [DH] tests/ref/fate/filter-pixfmts-pad
  • [DH] tests/ref/fate/filter-pixfmts-scale
  • [DH] tests/ref/fate/filter-pixfmts-transpose
  • [DH] tests/ref/fate/filter-pixfmts-vflip
  • [DH] tests/ref/fate/sws-pixdesc-query

• advanced ffmpeg compression controll

  27 décembre 2015, par Daniel Mahler

  I am using very aggressive video compression, eg -crf 51. I am using this for ’artistic’ effect, so what I am doing may not make sense from a normal video compression point of view.

  So far I have only been using very basic compression control using only the -crf or -b:v flags. The results look like ffmpeg divides images into square patches and the makes smooth approximations within the patches. This gives 2 control dimensions to the process : the patch size and the aggressiveness of the smoothing within the patches.

  It have found that ffmpeg uses both parameters to some extent, but there appears to be an absolute maximum patch size in pixels beyond which it will not go regardless of the frame size.
  After that it will only increase compression by reducing the detail within the patches.

  This is suboptimal for high resolution video, where this becomes equivalent to reducing the resolution. The problem is particularly noticeable on fractal like images which have large featureless region as well as regions of high detail.

  How can I tell ffmpeg to increase the maximum patch size and retain more detail within the patches ?