Recherche avancée

Sur d’autres sites (6102)

• ffmpeg problems with streaming mp4 over udp in local network

  28 novembre 2019, par AJ Cole

  I’m streaming mp4 video files (some of them are avi converted to mp4 with ffmpeg earlier) over udp://232.255.23.23:1234 from linux (embedded) with ffmpeg v3.4.2 to multiple linux (antix) machines that play the stream with MPV, all of this happens in local network so I expected it to work flawlessly, but unfortunately it doesn’t.

  Here are the original commands I tried to use :

  ffmpeg

  ffmpeg -re -i PATH_TO_FILE.mp4 -c copy -f mpegts udp://232.255.23.23:1234

  mpv

  mpv --no-config --geometry=[geo settings] --no-border udp://232.255.23.23:1234

  This seemed to woork good, however a problem appeared that on the displaying end, the stream is actually much longer than the streamed content itself. The mp4 files in total have 5 minutes 36 seconds, and mpv plays the entire stream loop in >= 6 minutes. I think it’s happening because of dropped frames, that mpv waits to recover or something and therefore extends the length of the actual content. This cannot work in my case, as I have a precise time gap for displaying the stream and it cannot be longer than the streamed content.
  All the content is made in 1680x800 resolution and is displayed on a screen with 1680x1050 resoltion (positioned with mpv geometry)

  It appears that using this command for mpv :

  mpv --no-config --framedrop=no --geometry=[geo settings] --no-border udp://232.255.23.23:1234

  made the duration correct, however this introduces huge artifacts in the videos sometimes.

  I read that using -re for streaming can cause these frame drops, so I tried putting a static number of fps for both file input and output stream, for example :

  ffmpeg -re -i PATH_TO_FILE.mp4 -c copy -r 25 -f mpegts udp://232.255.23.23:1234

  This reads the file at native framerate and outputs the stream at 25fps, and it appears to have the timing duration correct, but it also causes occasional articats and I think has worse qualit overall. Output from mpv when one of the artifacts happened :

  [ffmpeg/video] h264: cabac decode of qscale diff failed at 85 19
  
  [ffmpeg/video] h264: error while decoding MB 85 19, bytestream 85515

  I also tried using --untimed or --no-cache in mpv, but this causes stutters in the video

  I’m also getting requent Invalid video timestamp warnings in MPV, for example : Invalid video timestamp: 1.208333 -> -8.711667

  Playing in mpv without --no-config and with --untimed added also causes frequent artifacts :

  V: -00:00:00 / 00:00:00 Cache:  0s+266KB
  
  [ffmpeg/video] h264: Invalid NAL unit 8, skipping.
  
  V: -00:00:00 / 00:00:00 Cache:  0s+274KB
  
  [ffmpeg/video] h264: Reference 4 >= 4
  
  [ffmpeg/video] h264: error while decoding MB 6 0, bytestream 31474
  
  [ffmpeg/video] h264: error while decoding MB 78 49, bytestream -12
  
  V: 00:00:06 / 00:00:00 Cache:  5s+11KB
  
  Invalid video timestamp: 6.288333 -> -8.724933
  
  V: -00:00:05 / 00:00:00 Cache:  3s+0KB
  
  [ffmpeg/video] h264: Invalid NAL unit 8, skipping.
  
  [ffmpeg/video] h264: error while decoding MB 59 24, bytestream -27
  
  V: -00:00:04 / 00:00:00 Cache:  3s+0KB
  
  [ffmpeg/video] h264: Reference 4 >= 3
  
  [ffmpeg/video] h264: error while decoding MB 5 2, bytestream 13402
  
  V: -00:00:03 / 00:00:00 Cache:  2s+0KB
  
  [ffmpeg/video] h264: Reference 5 >= 4
  
  [ffmpeg/video] h264: error while decoding MB 51 21, bytestream 9415

  I tried playing the stream with ffplay and it also caused the videos to be "played" 20 seconds longer.
  Is there any way to keep the streaming duration intact and prevent those huge artifacts ? These aren’t any huge video files, they are few MB each, everything happens in local network so the latencies are minimal.

  Output from ffmpeg when streaming one of the files :

   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
  
    libswscale      4.  8.100 /  4.  8.100
  
    libswresample   2.  9.100 /  2.  9.100
  
  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'SDM.mp4':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 512
  
      compatible_brands: isomiso2avc1mp41
  
      encoder         : Lavf57.48.100
  
    Duration: 00:00:20.00, start: 0.000000, bitrate: 1883 kb/s
  
      Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1680x800 [SAR 1:1 DAR 21:10], 1880 kb/s, 24 fps, 24 tbr, 12288 tbn, 48 tbc (default)
  
      Metadata:
  
        handler_name    : VideoHandler
  
  Output #0, mpegts, to 'udp://232.255.23.23:1234':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 512
  
      compatible_brands: isomiso2avc1mp41
  
      encoder         : Lavf57.83.100
  
      Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1680x800 [SAR 1:1 DAR 21:10], q=2-31, 1880 kb/s, 24 fps, 24 tbr, 90k tbn, 25 tbc (default)
  
      Metadata:
  
        handler_name    : VideoHandler
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (copy)
  
  Press [q] to stop, [?] for help
  
  frame=  480 fps= 24 q=-1.0 Lsize=    5009kB time=00:00:19.87 bitrate=2064.7kbits/s speed=   1x
  
  video:4592kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 9.082929%

  Edit : all the files don’t contain any audio, so it should be even less traffic on the network

• How to discard corrupted frames in FFMPEG

  18 février 2020, par atokzz

  I’m running a command in FFMPEG that uses as input a stream video (Boroscope IP stream) and transform it to an .avi output.

  To do it, i ran the following command :

  ffmpeg -i tcp://192.168.10.123:7060 -c:v copy out.mov 

  When i ran this command, some errors appeared in console :

  https://prnt.sc/r3ranr

  Even with the errors, the video was encoded.

  What happens is that some frames has a strange quality problem, probably because of these errors :

  https://prnt.sc/r3rb1h

  Im not aiming to solve this problem, all I want to do is to discard these corrupt frames through some parameter in FFMPEG, because the frames with good quality are already enough for what I need to do.

  I tried to, instead of transforming the output to an .avi file, getting only the I-frames with the command :

  ffmpeg -i tcp://192.168.10.123:7060 "select='eq(pict_type,PICT_TYPE_I)'" -vsync vfr teste%04d.png

  With that command I was able to testify that the errors were directly connected to the frame with quality problems.

  I also ran a command that would discard the corrupted frames from the input (or at least that’s what I thought this command would do) :

  ffmpeg -err_detect bitstream -err_detect buffer -err_detect explode -err_detect aggressive -fflags discardcorrupt -i tcp://192.168.10.123:7060 -r 1 -an -f image2  snapshot-%03d.jpg

  But it didn’t work aswell.

  This is the first time I use FFMPEG, and I don’t know this framework and it’s commands very well.

  So, finally, is there ANY parameter in FFMPEG that can make me DISCARD all the frames where errors occurred (any type of error) ?

  Thanks.

• Creating GIF from QImages with ffmpeg

  21 mars 2020, par Sierra

  I would like to generate GIF from QImage, using ffmpeg - all of that programmatically (C++). I’m working with Qt 5.6 and the last build of ffmpeg (build git-0a9e781 (2016-06-10).

  I’m already able to convert these QImage in .mp4 and it works. I tried to use the same principle for the GIF, changing format pixel and codec. GIF is generated with two pictures (1 second each), in 15 FPS.

  ## INITIALIZATION
  
  #####################################################################
  
  
  
  // Filepath : "C:/Users/.../qt_temp.Jv7868.gif"  
  
  // Allocating an AVFormatContext for an output format...
  
  avformat_alloc_output_context2(formatContext, NULL, NULL, filepath);
  
  
  
  ...
  
  
  
  // Adding the video streams using the default format codecs and initializing the codecs.
  
  stream = avformat_new_stream(formatContext, *codec);
  
  
  
  AVCodecContext * codecContext = avcodec_alloc_context3(*codec);
  
  
  
  context->codec_id       = codecId;
  
  context->bit_rate       = 400000;
  
  ...
  
  context->pix_fmt        = AV_PIX_FMT_BGR8;
  
  
  
  ...
  
  
  
  // Opening the codec...
  
  avcodec_open2(codecContext, codec, NULL);
  
  
  
  ...
  
  
  
  frame = allocPicture(codecContext->width, codecContext->height, codecContext->pix_fmt);
  
  tmpFrame = allocPicture(codecContext->width, codecContext->height, AV_PIX_FMT_RGBA);
  
  
  
  ...
  
  
  
  avformat_write_header(formatContext, NULL);
  
  
  
  ## ADDING A NEW FRAME
  
  #####################################################################
  
  
  
  // Getting in parameter the QImage: newFrame(const QImage & image)
  
  const qint32 width  = image.width();
  
  const qint32 height = image.height();
  
  
  
  // Converting QImage into AVFrame
  
  for (qint32 y = 0; y < height; y++) {
  
      const uint8_t * scanline = image.scanLine(y);
  
  
  
      for (qint32 x = 0; x < width * 4; x++) {
  
          tmpFrame->data[0][y * tmpFrame->linesize[0] + x] = scanline[x];
  
      }
  
  }
  
  
  
  ...
  
  
  
  // Scaling...
  
  if (codec->pix_fmt != AV_PIX_FMT_BGRA) {
  
      if (!swsCtx) {
  
          swsCtx = sws_getContext(codec->width, codec->height,
  
                                  AV_PIX_FMT_BGRA,
  
                                  codec->width, codec->height,
  
                                  codec->pix_fmt,
  
                                  SWS_BICUBIC, NULL, NULL, NULL);
  
      }
  
  
  
      sws_scale(swsCtx,
  
                (const uint8_t * const *)tmpFrame->data,
  
                tmpFrame->linesize,
  
                0,
  
                codec->height,
  
                frame->data,
  
                frame->linesize);
  
  }
  
  frame->pts = nextPts++;
  
  
  
  ...
  
  
  
  int gotPacket = 0;
  
  AVPacket packet = {0};
  
  
  
  av_init_packet(&packet);
  
  avcodec_encode_video2(codec, &packet, frame, &gotPacket);
  
  
  
  if (gotPacket) {
  
      av_packet_rescale_ts(paket, *codec->time_base, stream->time_base);
  
      paket->stream_index = stream->index;
  
  
  
      av_interleaved_write_frame(formatContext, paket);
  
  }

  But when I’m trying to modify the video codec and pixel format to match with GIF specifications, I’m facing some issues.
  I tried several codecs such as AV_CODEC_ID_GIF and AV_CODEC_ID_RAWVIDEO but none of them seem to work. During the initialization phase, avcodec_open2() always returns such kind of errors :

  Specified pixel format rgb24 is invalid or not supported
  
  Could not open video codec:  gif

  EDIT 17/06/2016

  Digging a little bit more, avcodec_open2() returns -22 :

  #define EINVAL          22      /* Invalid argument */

  EDIT 22/06/2016

  Here are the flags used to compile ffmpeg :

  "FFmpeg/Libav configuration: --disable-static --enable-shared --enable-gpl --enable-version3 --disable-w32threads --enable-nvenc --enable-avisynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enable-iconv --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libfreetype --enable-libgme --enable-libgsm --enable-libilbc --enable-libmodplug --enable-libmfx --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-librtmp --enable-libschroedinger --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libtheora --enable-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-decklink --enable-zlib"

  Did I miss a crucial one for GIF ?

  EDIT 27/06/2016

  Thanks to Gwen, I have a first output : I setted the context->pix_fmt to AV_PIX_FMT_BGR8. Btw I’m still facing some issues with the generated GIF. It’s not playing and encoding appears to fail.

  GIF generated in command lines with ffmpeg (left) . . . GIF generated programmatically (right)
  
  

  It looks like some options are not defined... also may be a wrong conversion between QImage and AVFrame ? I updated the code above. It represents a lot of code, so I tried to stay short. Don’t hesitate to ask more details.

  End of EDIT

  I’m not really familiar with ffmpeg, any kind of help would be highly appreciated. Thank you.