Recherche avancée

Sur d’autres sites (9013)

• FFMPEG to create an MPEG-DASH stream with VP8

  16 septembre 2019, par Kenny Worden

  I’m trying to use FFMPEG to stream a live video feed from my webcam /dev/video0. Following scattered tutorials and scarce documentation (is this a known problem for the encoding community ?) I arrived at the following bash script :

  #!/bin/bash
  
  
  
  ffmpeg \
  
      -y \
  
      -f v4l2 \
  
          -i /dev/video0 \
  
          -s 640x480 \
  
          -input_format mjpeg \
  
          -r 24 \
  
      -map 0:0 \
  
      -pix_fmt yuv420p \
  
      -codec:v libvpx \
  
          -s 640x480 \
  
          -threads 4 \
  
          -b:v 50k \
  
          -tile-columns 4 \
  
          -frame-parallel 1 \
  
          -keyint_min 24 -g 24 \
  
      -f webm_chunk \
  
          -header "stream.hdr" \
  
          -chunk_start_index 1 \
  
      stream_%d.chk &
  
  
  
  sleep 2
  
  
  
  ffmpeg \
  
      -f webm_dash_manifest -live 1 \
  
      -i stream.hdr \
  
      -c copy \
  
      -map 0 \
  
      -f webm_dash_manifest -live 1 \
  
          -adaptation_sets "id=0,streams=0" \
  
          -chunk_start_index 1 \
  
          -chunk_duration_ms 1000 \
  
          -time_shift_buffer_depth 30000 \
  
          -minimum_update_period 60000 \
  
      stream_manifest.mpd

  When I run this script, my webcam light turns on, the stream.hdr and stream_manifest.mpd files are written, and chunks start to be created (i.e. stream_1.chk, stream_2.chk, etc...). However, FFMPEG throws the following error :

  Could not write header for output file #0 (incorrect codec parameters
   ?) : Invalid data found when processing input

  I will explain what I think I am doing with this script, and hopefully this will expose any errors in my thinking.

  First, we invoke FFMPEG to use Video for Linux 2 (v4l2) to read from my webcam (/dev/video0) of a resolution 640x480. The input format is mjpeg with a framerate of 24fps.

  I then declare that FFMPEG should "map" (copy) the video stream output by v4l2 to a file. I specify the pixel format (YUV420P) and use libvpx (VP8 encoding) to encode the video stream. I set the size to be 640x480, use 4 threads, set the bitrate to be 50kbps, do some magic with tile-columns and frame-parallel options, and set the I-frames to be 24 frames apart.

  I then create a stream.hdr file. The starting index is 1. This command continues to run infinitely until I kill it, grabbing new video from my webcam and outputting it into chunks.

  I then sleep for 2 seconds to give the previous command time to generate a header file.

  And that’s really it. The next invocation of FFMPEG simply creates the MPEG-DASH manifest file given the header generated in the previous step.

  So what’s going on ? Why can I not view the video in a web browser (I’m using Dash.js) ? I serve the manifest, header, and chunks on a Node.js server so that trivial issue is not the problem.

  ---

  Edit : Here is my full console output.

  ffmpeg version 3.0.7-0ubuntu0.16.10.1 Copyright (c) 2000-2017 the FFmpeg developers
  
    built with gcc 6.2.0 (Ubuntu 6.2.0-5ubuntu12) 20161005
  
    configuration: --prefix=/usr --extra-version=0ubuntu0.16.10.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --cc=cc --cxx=g++ --enable-gpl --enable-shared --disable-stripping --disable-decoder=libopenjpeg --disable-decoder=libschroedinger --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-libmodplug --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librubberband --enable-librtmp --enable-libschroedinger --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-libxvid --enable-libzvbi --enable-openal --enable-opengl --enable-x11grab --enable-libdc1394 --enable-libiec61883 --enable-libzmq --enable-frei0r --enable-chromaprint --enable-libx264
  
    libavutil      55. 17.103 / 55. 17.103
  
    libavcodec     57. 24.102 / 57. 24.102
  
    libavformat    57. 25.100 / 57. 25.100
  
    libavdevice    57.  0.101 / 57.  0.101
  
    libavfilter     6. 31.100 /  6. 31.100
  
    libavresample   3.  0.  0 /  3.  0.  0
  
    libswscale      4.  0.100 /  4.  0.100
  
    libswresample   2.  0.101 /  2.  0.101
  
    libpostproc    54.  0.100 / 54.  0.100
  
  [video4linux2,v4l2 @ 0x55847e244ea0] The driver changed the time per frame from 1/24 to 1/30
  
  [mjpeg @ 0x55847e245c00] Changing bps to 8
  
  Input #0, video4linux2,v4l2, from '/dev/video0':
  
    Duration: N/A, start: 64305.102081, bitrate: N/A
  
      Stream #0:0: Video: mjpeg, yuvj422p(pc, bt470bg/unknown/unknown), 640x480, -5 kb/s, 30 fps, 30 tbr, 1000k tbn, 1000k tbc
  
  Codec AVOption frame-parallel (Enable frame parallel decodability features) specified for output file #0 (stream_%d.chk) has not been used for any stream. The most likely reason is either wrong type (e.g. a video option with no video streams) or that it is a private option of some encoder which was not actually used for any stream.
  
  Codec AVOption tile-columns (Number of tile columns to use, log2) specified for output file #0 (stream_%d.chk) has not been used for any stream. The most likely reason is either wrong type (e.g. a video option with no video streams) or that it is a private option of some encoder which was not actually used for any stream.
  
  [swscaler @ 0x55847e24b720] deprecated pixel format used, make sure you did set range correctly
  
  [libvpx @ 0x55847e248a20] v1.5.0
  
  Output #0, webm_chunk, to 'stream_%d.chk':
  
    Metadata:
  
      encoder         : Lavf57.25.100
  
      Stream #0:0: Video: vp8 (libvpx), yuv420p, 640x480, q=-1--1, 50 kb/s, 30 fps, 30 tbn, 30 tbc
  
      Metadata:
  
        encoder         : Lavc57.24.102 libvpx
  
      Side data:
  
        unknown side data type 10 (24 bytes)
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (mjpeg (native) -> vp8 (libvpx))
  
  Press [q] to stop, [?] for help
  
  frame=   21 fps=0.0 q=0.0 size=N/A time=00:00:00.70 bitrate=N/A dup=5 drop=frame=   36 fps= 35 q=0.0 size=N/A time=00:00:01.20 bitrate=N/A dup=5 drop=frame=   51 fps= 33 q=0.0 size=N/A time=00:00:01.70 bitrate=N/A dup=5 drop=ffmpeg version 3.0.7-0ubuntu0.16.10.1 Copyright (c) 2000-2017 the FFmpeg developers
  
    built with gcc 6.2.0 (Ubuntu 6.2.0-5ubuntu12) 20161005
  
    configuration: --prefix=/usr --extra-version=0ubuntu0.16.10.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --cc=cc --cxx=g++ --enable-gpl --enable-shared --disable-stripping --disable-decoder=libopenjpeg --disable-decoder=libschroedinger --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-libmodplug --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librubberband --enable-librtmp --enable-libschroedinger --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-libxvid --enable-libzvbi --enable-openal --enable-opengl --enable-x11grab --enable-libdc1394 --enable-libiec61883 --enable-libzmq --enable-frei0r --enable-chromaprint --enable-libx264
  
    libavutil      55. 17.103 / 55. 17.103
  
    libavcodec     57. 24.102 / 57. 24.102
  
    libavformat    57. 25.100 / 57. 25.100
  
    libavdevice    57.  0.101 / 57.  0.101
  
    libavfilter     6. 31.100 /  6. 31.100
  
    libavresample   3.  0.  0 /  3.  0.  0
  
    libswscale      4.  0.100 /  4.  0.100
  
    libswresample   2.  0.101 /  2.  0.101
  
    libpostproc    54.  0.100 / 54.  0.100
  
  Input #0, webm_dash_manifest, from 'stream.hdr':
  
    Metadata:
  
      encoder         : Lavf57.25.100
  
    Duration: N/A, bitrate: N/A
  
      Stream #0:0: Video: vp8, yuv420p, 640x480, SAR 1:1 DAR 4:3, 30 fps, 30 tbr, 1k tbn, 1k tbc (default)
  
      Metadata:
  
        webm_dash_manifest_file_name: stream.hdr
  
        webm_dash_manifest_track_number: 1
  
  Output #0, webm_dash_manifest, to 'stream_manifest.mpd':
  
    Metadata:
  
      encoder         : Lavf57.25.100
  
      Stream #0:0: Video: vp8, yuv420p, 640x480 [SAR 1:1 DAR 4:3], q=2-31, 30 fps, 30 tbr, 1k tbn, 1k tbc (default)
  
      Metadata:
  
        webm_dash_manifest_file_name: stream.hdr
  
        webm_dash_manifest_track_number: 1
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (copy)
  
  Could not write header for output file #0 (incorrect codec parameters ?): Invalid data found when processing input
  
  frame=   67 fps= 33 q=0.0 size
  
  frame=   82 fps= 32 q=0.0 size=N/A time=00:00:02.73 bitrate=N/A dup=5 drop=
  
  frame=   97 fps= 32 q=0.0 size=N/A time=00:00:03.23 bitrate=N/A dup=5 drop=
  
  frame=  112 fps= 32 q=0.0 size=N/A time=00:00:03.73 bitrate=N/A dup=5 ...

• what FFMPEG performance settings to use for processing videos for the web

  19 juin 2015, par eAbi

  I have a few questions regarding usage of ffmpeg for processing videos for the web. I’m a beginner so please bear with me (although I read some docs on the internet)

  Performance

  • First of all, given the fact that FFMPEG utilizes all cores at 100%, what is the actual parallelism efficiency ?

  Let’s assume the following scenario. I have a video (fullHD, doesn’t matter what encoders / compression format was used to obtain the video) and I want to resize (downscale) to various sizes (e.g. 240px, 480px and 720px height) using mp4 format (thus using libx264 with aac codecs).

  Using ffmpeg, I see that all of my laptop’s cores (8) are used at 100% and I was wondering what scenarios can improve the overall performance of the whole processing task. So this leads us to basically 2 scenarios : Assuming the video mentioned above as input, for obtaining the 3 output videos (@ 240px, 480px and 720px height sizes), we :

  1. Process input video and obtain 1 output video at a time, and let all the cores work at the same time at 100% ;
  2. Process the video to obtain all output videos in parallel, by bounding each output video to a single processor core which’ll work at 100% ;

  So the question is actually reduced to the parallelism efficiency of the ffmpeg program.

  This means that letting ffmpeg process the task procVideo - which takes 1 input video to produce 1 single output video (transcoding/downscaling and so on) - on N processor cores doesn’t mean it finish the task N times faster than letting it run the same task bound to a single core. So if the efficiency is smaller than 100%, it’s better to have N procVideo tasks in parallel, each bound to a single core, rather than doing the task sequentially for each output video.

  ---

  Codecs

  Other than the above performance problem, the usage of codecs bugs me. I am trying to obtain mp4 videos because of the wide implementation of the format in html5 browsers.

  So having a video as input in any format, I want to convert it to mp4. So I’m using libx264 codec with aac.

  • Use libx264, x264 or h264 for video encoding/decoding ?
  • Use libfdk_aac, libaacplus or aac for audio encoding/decoding to aac ?

  Also, I would like to know what are the licesing fees for each of the above codec, as the online resources on these are quite limited / hard to understand.

  If anyone could shed some light on those questions, I would really be grateful ! Thanks for your time !

• matplotlib funcanimation save issue

  20 avril 2015, par Richie Abraham

  I have been trying some animation recently using matplotlib animations. It has been going great, i create an ffmpeg writer and save it as a video file. However i face an issue whenever the function that FuncAnimation calls returns more than one object.

  Below is a small snippet of my code base. When I return both im0 and im1, the video file created only has im1 , although the plt.show command works as expected ( showing both the videos ). If i return just a single im0, then it works as expected. IT also works as expected if i return both im0 and im1 with alpha=0.5.

  Can anyone shed some light on what is happening underneath the hood ?

  fig, ax = plt.subplots(1)
  
  def animate(i):
  
      im0=ax.imshow(np.ma.masked_array(imgl[i][:,:,0], mask=get_blob(i)),cmap='cubehelix')
  
  
  
      im1=ax.imshow(imgl[(i-100)%len(imgl)][:,:,0],cmap='cubehelix')
  
  
  
      return [im1,im0]
  
  
  
  
  
  
  
  ani = animation.FuncAnimation(fig, animate, frames=200,
  
                                interval=10, blit=True,repeat=False)
  
  ani.save('ps.mp4', writer=writer)
  
  plt.show()