Recherche avancée

Sur d’autres sites (6950)

• Video too fast FFmpeg

  22 novembre 2012, par Spamdark

  I am having an issue again with ffmpeg, I'm a newbie with ffmpeg, and I can't find a good tutorial up to date...

  This time, when I play a video with ffmpeg, it plays too fast, ffmpeg is ignoring the FPS, I don't want to handle that with a thread sleep, because the videos have differents FPS's.

  I created a thread, there you can find the loop :

  AVPacket framepacket;
  
  
  
  while(av_read_frame(formatContext,&framepacket)>= 0){
  
      pausecontrol.lock();
  
  
  
      // Is it a video or audio frame¿?
  
      if(framepacket.stream_index==gotVideoCodec){
  
          int framereaded;
  
          // Video? Ok
  
          avcodec_decode_video2(videoCodecContext,videoFrame,&framereaded,&framepacket);
  
          // Yeah, did we get it?
  
          if(framereaded && doit){
  
              AVRational millisecondbase = {1,1000};
  
              int f_number = framepacket.dts;
  
              int f_time = av_rescale_q(framepacket.dts,formatContext->streams[gotVideoCodec]->time_base,millisecondbase);
  
              currentTime=f_time;
  
              currentFrameNumber=f_number;
  
  
  
              int stWidth = videoCodecContext->width;
  
              int stHeight = videoCodecContext->height;
  
              SwsContext *ctx = sws_getContext(stWidth, stHeight, videoCodecContext->pix_fmt, stWidth,
  
              stHeight, PIX_FMT_RGB24, SWS_BICUBIC, NULL, NULL, NULL);
  
              if(ctx!=0){
  
              sws_scale(ctx,videoFrame->data,videoFrame->linesize,0,videoCodecContext->height,videoFrameRGB->data,videoFrameRGB->linesize);
  
              QImage framecapsule=QImage(stWidth,stHeight,QImage::Format_RGB888);
  
  
  
              for(int y=0;ydata[0]+y*videoFrameRGB->linesize[0],stWidth*3);
  
              }
  
              emit newFrameReady(framecapsule);
  
              sws_freeContext(ctx);
  
              }
  
  
  
          }
  
      }
  
      if(framepacket.stream_index==gotAudioCodec){
  
          // Audio? Ok
  
      }
  
      pausecontrol.unlock();
  
      av_free_packet(&framepacket);
  
  }

  Any Idea ?

• ffmpeg + ffserver : "Broken ffmpeg default settings detected"

  18 octobre 2012, par Chris Nolet

  I'm just trying to connect ffmpeg to ffserver and stream rawvideo.

  I keep getting the error : broken ffmpeg default settings detected from libx264 and then Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height from ffmpeg before it exits.

  I'm launching ffmpeg with the command : ffmpeg -f x11grab -s 320x480 -r 10 -i :0.0 -tune zerolatency http://localhost:8090/feed1.ffm

  My ffserver.conf file (for ffserver) looks like this :

  Port 8090
  
  BindAddress 0.0.0.0
  
  MaxHTTPConnections 2000
  
  MaxClients 1000
  
  MaxBandwidth 1000
  
  CustomLog -
  
  NoDaemon
  
  
  
  <feed>
  
    ACL allow 127.0.0.1
  
  </feed>
  
  
  
  <stream>
  
    Feed feed1.ffm
  
    Format asf
  
  
  
    NoAudio
  
  
  
    VideoBitRate 128
  
    VideoBufferSize 400
  
    VideoFrameRate 24
  
    VideoSize 320x480
  
  
  
    VideoGopSize 12
  
  
  
    VideoQMin 1
  
    VideoQMax 31
  
  
  
    VideoCodec libx264
  
  </stream>
  
  
  
  <stream>
  
    Format status
  
  </stream>

  And the full output is :

  ffmpeg version N-45614-g364c60b Copyright (c) 2000-2012 the FFmpeg developers
  
    built on Oct 17 2012 04:34:04 with Apple clang version 4.1 (tags/Apple/clang-421.11.65) (based on LLVM 3.1svn)
  
    configuration: --enable-shared --enable-libx264 --enable-libmp3lame --enable-x11grab --enable-gpl --enable-version3 --enable-nonfree --enable-hardcoded-tables --cc=/usr/bin/clang --host-cflags='-Os -w -pipe -march=native -Qunused-arguments -mmacosx-version-min=10.7' --extra-cflags='-x objective-c' --extra-ldflags='-framework Foundation -framework Cocoa -framework CoreServices -framework ApplicationServices -lobjc'
  
    libavutil      51. 76.100 / 51. 76.100
  
    libavcodec     54. 66.100 / 54. 66.100
  
    libavformat    54. 32.101 / 54. 32.101
  
    libavdevice    54.  3.100 / 54.  3.100
  
    libavfilter     3. 19.103 /  3. 19.103
  
    libswscale      2.  1.101 /  2.  1.101
  
    libswresample   0. 16.100 /  0. 16.100
  
    libpostproc    52.  1.100 / 52.  1.100
  
  [x11grab @ 0x7f87dc01e200] device: :0.0 -> display: :0.0 x: 0 y: 0 width: 320 height: 480
  
  [x11grab @ 0x7f87dc01e200] Estimating duration from bitrate, this may be inaccurate
  
  Input #0, x11grab, from ':0.0':
  
    Duration: N/A, start: 1350517708.386699, bitrate: 49152 kb/s
  
      Stream #0:0: Video: rawvideo (BGRA / 0x41524742), bgra, 320x480, 49152 kb/s, 10 tbr, 1000k tbn, 10 tbc
  
  [tcp @ 0x7f87dc804120] TCP connection to localhost:8090 failed: Connection refused
  
  [tcp @ 0x7f87dc804b20] TCP connection to localhost:8090 failed: Connection refused
  
  [libx264 @ 0x7f87dd801000] broken ffmpeg default settings detected
  
  [libx264 @ 0x7f87dd801000] use an encoding preset (e.g. -vpre medium)
  
  [libx264 @ 0x7f87dd801000] preset usage: -vpre <speed> -vpre <profile>
  
  [libx264 @ 0x7f87dd801000] speed presets are listed in x264 --help
  
  [libx264 @ 0x7f87dd801000] profile is optional; x264 defaults to high
  
  Output #0, ffm, to &#39;http://localhost:8090/feed1.ffm&#39;:
  
    Metadata:
  
      creation_time   : now
  
      Stream #0:0: Video: h264, yuv420p, 160x128, q=2-31, 128 kb/s, 1000k tbn, 10 tbc
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (rawvideo -> libx264)
  
  Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height
  
  </profile></speed>

  Any help much appreciated :)

• Compute PTS and DTS correctly to sync audio and video ffmpeg C++

  14 août 2015, par Kaidul Islam

  I am trying to mux H264 encoded data and G711 PCM data into mov multimedia container. I am creating AVPacket from encoded data and initially the PTS and DTS value of video/audio frames is equivalent to AV_NOPTS_VALUE. So I calculated the DTS using current time information. My code -

  bool AudioVideoRecorder::WriteVideo(const unsigned char *pData, size_t iDataSize, bool const bIFrame) {
  
      .....................................
  
      .....................................
  
      .....................................
  
      AVPacket pkt = {0};
  
      av_init_packet(&pkt);
  
      int64_t dts = av_gettime();
  
      dts = av_rescale_q(dts, (AVRational){1, 1000000}, m_pVideoStream->time_base);
  
      int duration = 90000 / VIDEO_FRAME_RATE;
  
      if(m_prevVideoDts > 0LL) {
  
          duration = dts - m_prevVideoDts;
  
      }
  
      m_prevVideoDts = dts;
  
  
  
      pkt.pts = AV_NOPTS_VALUE;
  
      pkt.dts = m_currVideoDts;
  
      m_currVideoDts += duration;
  
      pkt.duration = duration;
  
      if(bIFrame) {
  
          pkt.flags |= AV_PKT_FLAG_KEY;
  
      }
  
      pkt.stream_index = m_pVideoStream->index;
  
      pkt.data = (uint8_t*) pData;
  
      pkt.size = iDataSize;
  
  
  
      int ret = av_interleaved_write_frame(m_pFormatCtx, &pkt);
  
  
  
      if(ret < 0) {
  
          LogErr("Writing video frame failed.");
  
          return false;
  
      }
  
  
  
      Log("Writing video frame done.");
  
  
  
      av_free_packet(&pkt);
  
      return true;
  
  }
  
  
  
  bool AudioVideoRecorder::WriteAudio(const unsigned char *pEncodedData, size_t iDataSize) {
  
      .................................
  
      .................................
  
      .................................
  
      AVPacket pkt = {0};
  
      av_init_packet(&pkt);
  
  
  
      int64_t dts = av_gettime();
  
      dts = av_rescale_q(dts, (AVRational){1, 1000000}, (AVRational){1, 90000});
  
      int duration = AUDIO_STREAM_DURATION; // 20
  
      if(m_prevAudioDts > 0LL) {
  
          duration = dts - m_prevAudioDts;
  
      }
  
      m_prevAudioDts = dts;
  
      pkt.pts = AV_NOPTS_VALUE;
  
      pkt.dts = m_currAudioDts;
  
      m_currAudioDts += duration;
  
      pkt.duration = duration;
  
  
  
      pkt.stream_index = m_pAudioStream->index;
  
      pkt.flags |= AV_PKT_FLAG_KEY;
  
      pkt.data = (uint8_t*) pEncodedData;
  
      pkt.size = iDataSize;
  
  
  
      int ret = av_interleaved_write_frame(m_pFormatCtx, &pkt);
  
      if(ret < 0) {
  
          LogErr("Writing audio frame failed: %d", ret);
  
          return false;
  
      }
  
  
  
      Log("Writing audio frame done.");
  
  
  
      av_free_packet(&pkt);
  
      return true;
  
  }

  And I added stream like this -

  AVStream* AudioVideoRecorder::AddMediaStream(enum AVCodecID codecID) {
  
      ................................
  
      .................................   
  
      pStream = avformat_new_stream(m_pFormatCtx, codec);
  
      if (!pStream) {
  
          LogErr("Could not allocate stream.");
  
          return NULL;
  
      }
  
      pStream->id = m_pFormatCtx->nb_streams - 1;
  
      pCodecCtx = pStream->codec;
  
      pCodecCtx->codec_id = codecID;
  
  
  
      switch(codec->type) {
  
      case AVMEDIA_TYPE_VIDEO:
  
          pCodecCtx->bit_rate = VIDEO_BIT_RATE;
  
          pCodecCtx->width = PICTURE_WIDTH;
  
          pCodecCtx->height = PICTURE_HEIGHT;
  
          pStream->time_base = (AVRational){1, 90000};
  
          pStream->avg_frame_rate = (AVRational){90000, 1};
  
          pStream->r_frame_rate = (AVRational){90000, 1}; // though the frame rate is variable and around 15 fps
  
          pCodecCtx->pix_fmt = STREAM_PIX_FMT;
  
          m_pVideoStream = pStream;
  
          break;
  
  
  
      case AVMEDIA_TYPE_AUDIO:
  
          pCodecCtx->sample_fmt = AV_SAMPLE_FMT_S16;
  
          pCodecCtx->bit_rate = AUDIO_BIT_RATE;
  
          pCodecCtx->sample_rate = AUDIO_SAMPLE_RATE;
  
          pCodecCtx->channels = 1;
  
          m_pAudioStream = pStream;
  
          break;
  
  
  
      default:
  
          break;
  
      }
  
  
  
      /* Some formats want stream headers to be separate. */
  
      if (m_pOutputFmt->flags & AVFMT_GLOBALHEADER)
  
          m_pFormatCtx->flags |= CODEC_FLAG_GLOBAL_HEADER;
  
  
  
      return pStream;
  
  }

  There are several problems with this calculation :

  1. The video is laggy and lags behind than audio increasingly with time.

  2. Suppose, an audio frame is received (WriteAudio(..)) little lately like 3 seconds, then the late frame should be started playing with 3 second delay, but it’s not. The delayed frame is played consecutively with previous frame.

  3. Sometimes I recorded for 40 seconds but the file duration is much like 2 minutes, but audio/video is played only few moments like 40 seconds and rest of the file contains nothing and seekbar jumps at en immediately after 40 seconds (tested in VLC).

  EDIT :

  According to Ronald S. Bultje’s suggestion, what I’ve understand :

  m_pAudioStream->time_base = (AVRational){1, 9000}; // actually no need to set as 9000 is already default value for audio as you said
  
  m_pVideoStream->time_base = (AVRational){1, 9000};

  should be set as now both audio and video streams are now in same time base units.

  And for video :

  ...................
  
  ...................
  
  
  
  int64_t dts = av_gettime(); // get current time in microseconds
  
  dts *= 9000; 
  
  dts /= 1000000; // 1 second = 10^6 microseconds
  
  pkt.pts = AV_NOPTS_VALUE; // is it okay?
  
  pkt.dts = dts;
  
  // and no need to set pkt.duration, right?

  And for audio : (exactly same as video, right ?)

  ...................
  
  ...................
  
  
  
  int64_t dts = av_gettime(); // get current time in microseconds
  
  dts *= 9000; 
  
  dts /= 1000000; // 1 second = 10^6 microseconds
  
  pkt.pts = AV_NOPTS_VALUE; // is it okay?
  
  pkt.dts = dts;
  
  // and no need to set pkt.duration, right?

  And I think they are now like sharing same currDts, right ? Please correct me if I am wrong anywhere or missing anything.

  Also, if I want to use video stream time base as (AVRational){1, frameRate} and audio stream time base as (AVRational){1, sampleRate}, how the correct code should look like ?

  EDIT 2.0 :

  m_pAudioStream->time_base = (AVRational){1, VIDEO_FRAME_RATE};
  
  m_pVideoStream->time_base = (AVRational){1, VIDEO_FRAME_RATE};

  And

  bool AudioVideoRecorder::WriteAudio(const unsigned char *pEncodedData, size_t iDataSize) {
  
      ...........................
  
      ......................
  
      AVPacket pkt = {0};
  
      av_init_packet(&pkt);
  
  
  
      int64_t dts = av_gettime() / 1000; // convert into millisecond
  
      dts = dts * VIDEO_FRAME_RATE;
  
      if(m_dtsOffset < 0) {
  
          m_dtsOffset = dts;
  
      }
  
  
  
      pkt.pts = AV_NOPTS_VALUE;
  
      pkt.dts = (dts - m_dtsOffset);
  
  
  
      pkt.stream_index = m_pAudioStream->index;
  
      pkt.flags |= AV_PKT_FLAG_KEY;
  
      pkt.data = (uint8_t*) pEncodedData;
  
      pkt.size = iDataSize;
  
  
  
      int ret = av_interleaved_write_frame(m_pFormatCtx, &pkt);
  
      if(ret < 0) {
  
          LogErr("Writing audio frame failed: %d", ret);
  
          return false;
  
      }
  
  
  
      Log("Writing audio frame done.");
  
  
  
      av_free_packet(&pkt);
  
      return true;
  
  }
  
  
  
  bool AudioVideoRecorder::WriteVideo(const unsigned char *pData, size_t iDataSize, bool const bIFrame) {
  
      ........................................
  
      .................................
  
      AVPacket pkt = {0};
  
      av_init_packet(&pkt);
  
      int64_t dts = av_gettime() / 1000;
  
      dts = dts * VIDEO_FRAME_RATE;
  
      if(m_dtsOffset < 0) {
  
          m_dtsOffset = dts;
  
      }
  
      pkt.pts = AV_NOPTS_VALUE;
  
      pkt.dts = (dts - m_dtsOffset);
  
  
  
      if(bIFrame) {
  
          pkt.flags |= AV_PKT_FLAG_KEY;
  
      }
  
      pkt.stream_index = m_pVideoStream->index;
  
      pkt.data = (uint8_t*) pData;
  
      pkt.size = iDataSize;
  
  
  
      int ret = av_interleaved_write_frame(m_pFormatCtx, &pkt);
  
  
  
      if(ret < 0) {
  
          LogErr("Writing video frame failed.");
  
          return false;
  
      }
  
  
  
      Log("Writing video frame done.");
  
  
  
      av_free_packet(&pkt);
  
      return true;
  
  }

  Is the last change okay ? The video and audio seems synced. Only problem is - the audio is played without the delay regardless the packet arrived in delay.
  Like -

  packet arrival : 1 2 3 4... (then next frame arrived after 3 sec) .. 5

  audio played : 1 2 3 4 (no delay) 5

  EDIT 3.0 :

  zeroed audio sample data :

  AVFrame* pSilentData;
  
  pSilentData = av_frame_alloc();
  
  memset(&pSilentData->data[0], 0, iDataSize);
  
  
  
  pkt.data = (uint8_t*) pSilentData;
  
  pkt.size = iDataSize;
  
  
  
  av_freep(&pSilentData->data[0]);
  
  av_frame_free(&pSilentData);

  Is this okay ? But after writing this into file container, there are dot dot noise during playing the media. Whats the problem ?

  EDIT 4.0 :

  Well, For µ-Law audio the zero value is represented as 0xff. So -

  memset(&pSilentData->data[0], 0xff, iDataSize);

  solve my problem.