Recherche avancée

Sur d’autres sites (8969)

• Android MediaRecorder setCaptureRate() and video playback speed

  7 novembre 2013, par spitzanator

  I've got a MediaRecorder recording video, and I'm very confused by the effect of setCaptureRate().

  Specifically, I prepare my MediaRecorder as follows :

  mMediaRecorder = new MediaRecorder();
  
  mCamera.stopPreview();
  
  mCamera.unlock();
  
  mMediaRecorder.setCamera(mCamera);
  
  mMediaRecorder.setVideoSource(MediaRecorder.VideoSource.CAMERA);
  
  mMediaRecorder.setProfile(CamcorderProfile.QUALITY_TIME_LAPSE_480P);
  
  mMediaRecorder.setCaptureRate(30f);
  
  mMediaRecorder.setOrientationHint(270);
  
  mMediaRecorder.setOutputFile(...);
  
  mMediaRecorder.setPreviewDisplay(...);
  
  mMediaRecorder.prepare();

  I record for five seconds (with a CountDownTimer, but that's irrelevant), and this is the file that gets generated :

  $ ffmpeg -i ~/CaptureRate30fps.mp4 
  
  ...
  
  Seems stream 0 codec frame rate differs from container frame rate: 180000.00 (180000/1) -> 30.00 (30/1)
  
  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/home/mspitz/CaptureRate30fps.mp4':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 0
  
      compatible_brands: isom3gp4
  
      creation_time   : 2013-06-04 00:52:00
  
    Duration: 00:00:02.59, start: 0.000000, bitrate: 5238 kb/s
  
      Stream #0.0(eng): Video: h264 (Baseline), yuv420p, 720x480, 5235 kb/s, PAR 65536:65536 DAR 3:2, 30 fps, 30 tbr, 90k tbn, 180k tbc
  
      Metadata:
  
        creation_time   : 2013-06-04 00:52:00

  Note that the Duration is just about 3 seconds. The video also plays much faster, as if it were 5 seconds of video crammed into 3.

  Now, if I record by preparing my mediaRecorder exactly as above, but subtracting the setCaptureRate(30f) line, I get a file like this :

  $ ffmpeg -i ~/NoSetCaptureRate.mp4 
  
  ...
  
  Seems stream 0 codec frame rate differs from container frame rate: 180000.00 (180000/1) -> 90000.00 (180000/2)
  
  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/home/mspitz/NoSetCaptureRate.mp4':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 0
  
      compatible_brands: isom3gp4
  
      creation_time   : 2013-06-04 00:50:41
  
    Duration: 00:00:04.87, start: 0.000000, bitrate: 2803 kb/s
  
      Stream #0.0(eng): Video: h264 (Baseline), yuv420p, 720x480, 2801 kb/s, PAR 65536:65536 DAR 3:2, 16.01 fps, 90k tbr, 90k tbn, 180k tbc
  
      Metadata:
  
        creation_time   : 2013-06-04 00:50:41

  Note that the Duration is as expected, about 5 seconds. The video also plays at a normal speed.

  I'm using setCaptureRate(30f) because 30 frames per second is the value of my CamcorderProfile's videoFrameRate. On my Galaxy Nexus S2 (4.2.1), omitting setCaptureRate() is fine, but when I tested on a Galaxy Nexus S3 (4.1.1), omitting setCaptureRate() results in the ever-helpful "start failed -22" error when I called mMediaRecorder.start().

  So, what am I missing ? I thought that the capture rate and the video frame rate were independent, but it's clear that they're not. Is there a way to determine programmatically what I need to set the capture rate at in order to determine that my video plays back at 1x speed ?

• ffplay : use start_time as next audio pts on flush when seeking is not supported

  13 juillet 2013, par Marton Balint

  ffplay : use start_time as next audio pts on flush when seeking is not supported
  Theoretically using start_time should also work if seeking is available and we
  
  could determine that the next packet after a flush packet is the first packet
  
  of a stream, but I could not think of an easy and clean way to do that, that is
  
  why I sticked to the no seeking available condition for now.
  Fixes ticket #2647.
  Signed-off-by : Marton Balint <cus@passwd.hu>
  

  • [DH] ffplay.c

• FFmpeg avcodec_decode_video2 decode RTSP H264 HD-video packet to video picture with error

  29 mai 2018, par Nguyen Ba Thi

  I used FFmpeg library version 4.0 to have simple C++ program, in witch is a thread to receive RTSP H264 video data from IP-camera and display it in program window.

  Code of this thread is follow :

  DWORD WINAPI GrabbProcess(LPVOID lpParam)
  
  // Grabbing thread
  
  {
  
    DWORD i;
  
    int ret = 0, nPacket=0;
  
    FILE *pktFile;
  
    // Open video file
  
    pFormatCtx = avformat_alloc_context();
  
    if(avformat_open_input(&pFormatCtx, nameVideoStream, NULL, NULL)!=0)
  
        fGrabb=-1; // Couldn't open file
  
    else
  
    // Retrieve stream information
  
    if(avformat_find_stream_info(pFormatCtx, NULL)<0)
  
        fGrabb=-2; // Couldn't find stream information
  
    else
  
    {
  
        // Find the first video stream
  
        videoStream=-1;
  
        for(i=0; inb_streams; i++)
  
          if(pFormatCtx->streams[i]->codec->codec_type==AVMEDIA_TYPE_VIDEO)
  
          {
  
            videoStream=i;
  
            break;
  
          }
  
        if(videoStream==-1)
  
            fGrabb=-3; // Didn't find a video stream
  
        else
  
        {
  
            // Get a pointer to the codec context for the video stream
  
            pCodecCtxOrig=pFormatCtx->streams[videoStream]->codec;
  
            // Find the decoder for the video stream
  
            pCodec=avcodec_find_decoder(pCodecCtxOrig->codec_id);
  
            if(pCodec==NULL)
  
                fGrabb=-4; // Codec not found
  
            else
  
            {
  
                // Copy context
  
                pCodecCtx = avcodec_alloc_context3(pCodec);
  
                if(avcodec_copy_context(pCodecCtx, pCodecCtxOrig) != 0)
  
                    fGrabb=-5; // Error copying codec context
  
                else
  
                {
  
                    // Open codec
  
                    if(avcodec_open2(pCodecCtx, pCodec, NULL)<0)
  
                        fGrabb=-6; // Could not open codec
  
                    else
  
                    // Allocate video frame for input
  
                    pFrame=av_frame_alloc();
  
                    // Determine required buffer size and allocate buffer
  
                    numBytes=avpicture_get_size(pCodecCtx->pix_fmt, pCodecCtx->width,
  
                        pCodecCtx->height);
  
                    buffer=(uint8_t *)av_malloc(numBytes*sizeof(uint8_t));
  
                    // Assign appropriate parts of buffer to image planes in pFrame
  
                    // Note that pFrame is an AVFrame, but AVFrame is a superset
  
                    // of AVPicture
  
                    avpicture_fill((AVPicture *)pFrame, buffer, pCodecCtx->pix_fmt,
  
                        pCodecCtx->width, pCodecCtx->height);
  
  
  
                    // Allocate video frame for display
  
                    pFrameRGB=av_frame_alloc();
  
                    // Determine required buffer size and allocate buffer
  
                    numBytes=avpicture_get_size(AV_PIX_FMT_RGB24, pCodecCtx->width,
  
                        pCodecCtx->height);
  
                    bufferRGB=(uint8_t *)av_malloc(numBytes*sizeof(uint8_t));
  
                    // Assign appropriate parts of buffer to image planes in pFrameRGB
  
                    // Note that pFrameRGB is an AVFrame, but AVFrame is a superset
  
                    // of AVPicture
  
                    avpicture_fill((AVPicture *)pFrameRGB, bufferRGB, AV_PIX_FMT_RGB24,
  
                        pCodecCtx->width, pCodecCtx->height);
  
                    // initialize SWS context for software scaling to FMT_RGB24
  
                    sws_ctx_to_RGB = sws_getContext(pCodecCtx->width,
  
                        pCodecCtx->height,
  
                        pCodecCtx->pix_fmt,
  
                        pCodecCtx->width,
  
                        pCodecCtx->height,
  
                        AV_PIX_FMT_RGB24,
  
                        SWS_BILINEAR,
  
                        NULL,
  
                        NULL,
  
                        NULL);
  
  
  
                    // Allocate video frame (grayscale YUV420P) for processing
  
                    pFrameYUV=av_frame_alloc();
  
                    // Determine required buffer size and allocate buffer
  
                    numBytes=avpicture_get_size(AV_PIX_FMT_YUV420P, pCodecCtx->width,
  
                        pCodecCtx->height);
  
                    bufferYUV=(uint8_t *)av_malloc(numBytes*sizeof(uint8_t));
  
                    // Assign appropriate parts of buffer to image planes in pFrameYUV
  
                    // Note that pFrameYUV is an AVFrame, but AVFrame is a superset
  
                    // of AVPicture
  
                    avpicture_fill((AVPicture *)pFrameYUV, bufferYUV, AV_PIX_FMT_YUV420P,
  
                        pCodecCtx->width, pCodecCtx->height);
  
                    // initialize SWS context for software scaling to FMT_YUV420P
  
                    sws_ctx_to_YUV = sws_getContext(pCodecCtx->width,
  
                        pCodecCtx->height,
  
                        pCodecCtx->pix_fmt,
  
                        pCodecCtx->width,
  
                        pCodecCtx->height,
  
                        AV_PIX_FMT_YUV420P,
  
                        SWS_BILINEAR,
  
                        NULL,
  
                        NULL,
  
                        NULL);
  
                  RealBsqHdr.biWidth = pCodecCtx->width;
  
                  RealBsqHdr.biHeight = -pCodecCtx->height;
  
                }
  
            }
  
        }
  
    }
  
    while ((fGrabb==1)||(fGrabb==100))
  
    {
  
        // Grabb a frame
  
        if (av_read_frame(pFormatCtx, &packet) >= 0)
  
        {
  
          // Is this a packet from the video stream?
  
          if(packet.stream_index==videoStream)
  
          {
  
              // Decode video frame
  
              int len = avcodec_decode_video2(pCodecCtx, pFrame, &frameFinished, &packet);
  
              nPacket++;
  
              // Did we get a video frame?
  
              if(frameFinished)
  
              {
  
                  // Convert the image from its native format to YUV
  
                  sws_scale(sws_ctx_to_YUV, (uint8_t const * const *)pFrame->data,
  
                      pFrame->linesize, 0, pCodecCtx->height,
  
                      pFrameYUV->data, pFrameYUV->linesize);
  
                  // Convert the image from its native format to RGB
  
                  sws_scale(sws_ctx_to_RGB, (uint8_t const * const *)pFrame->data,
  
                      pFrame->linesize, 0, pCodecCtx->height,
  
                      pFrameRGB->data, pFrameRGB->linesize);
  
                  HDC hdc=GetDC(hWndM);
  
                  SetDIBitsToDevice(hdc, 0, 0, pCodecCtx->width, pCodecCtx->height,
  
                      0, 0, 0, pCodecCtx->height,pFrameRGB->data[0], (LPBITMAPINFO)&RealBsqHdr, DIB_RGB_COLORS);
  
                  ReleaseDC(hWndM,hdc);
  
                  av_frame_unref(pFrame);
  
              }
  
          }
  
          // Free the packet that was allocated by av_read_frame
  
          av_free_packet(&packet);
  
        }
  
     }
  
     // Free the org frame
  
    av_frame_free(&pFrame);
  
    // Free the RGB frame
  
    av_frame_free(&pFrameRGB);
  
    // Free the YUV frame
  
    av_frame_free(&pFrameYUV);
  
  
  
    // Close the codec
  
    avcodec_close(pCodecCtx);
  
    avcodec_close(pCodecCtxOrig);
  
  
  
    // Close the video file
  
    avformat_close_input(&pFormatCtx);
  
    avformat_free_context(pFormatCtx);
  
  
  
    if (fGrabb==1)
  
        sprintf(tmpstr,"Grabbing Completed %d frames", nCntTotal);
  
    else if (fGrabb==2)
  
        sprintf(tmpstr,"User break on %d frames", nCntTotal);
  
    else if (fGrabb==3)
  
        sprintf(tmpstr,"Can't Grabb at frame %d", nCntTotal);
  
    else if (fGrabb==-1)
  
        sprintf(tmpstr,"Couldn't open file");
  
    else if (fGrabb==-2)
  
        sprintf(tmpstr,"Couldn't find stream information");
  
    else if (fGrabb==-3)
  
        sprintf(tmpstr,"Didn't find a video stream");
  
    else if (fGrabb==-4)
  
        sprintf(tmpstr,"Codec not found");
  
    else if (fGrabb==-5)
  
        sprintf(tmpstr,"Error copying codec context");
  
    else if (fGrabb==-6)
  
        sprintf(tmpstr,"Could not open codec");
  
    i=(UINT) fGrabb;
  
    fGrabb=0;
  
    SetWindowText(hWndM,tmpstr);
  
    ExitThread(i);
  
    return 0;
  
  }
  
  // End Grabbing thread  

  When program receive RTSP H264 video data with resolution 704x576 then decoded video pictures are OK. When receive RTSP H264 HD-video data with resolution 1280x720 it look like that first video picture is decoded OK and then video pictures are decoded but always with some error.

  Please help me to fix this problem !

  Here is problems brief :
  I have an IP camera model HI3518E_50H10L_S39 (product of China).
  Camera can provide H264 video stream both at resolution 704x576 (with RTSP URI "rtsp ://192.168.1.18:554/user=admin_password=tlJwpbo6_channel=1_stream=1.sdp ?real_stream") or 1280x720 (with RTSP URI "rtsp ://192.168.1.18:554/user=admin_password=tlJwpbo6_channel=1_stream=0.sdp ?real_stream").
  Using FFplay utility I can access and display them with good picture quality.
  For testing of grabbing from this camera, I have a simple (above mentioned) program in VC-2005. In "Grabbing thread" program use FFmpeg library version 4.0 for opening camera RTSP stream, retrieve stream information, find the first video stream... and prepare some variables.
  Center of this thread is loop : Grab a frame (function av_read_frame) - Decode it if it’s video (function avcodec_decode_video2) - Convert to RGB format (function sws_scale) - Display to program window (GDI function SetDIBitsToDevice).
  When proram run with camera RTSP stream at resolution 704x576, I have good video picture. Here is a sample :
  704x576 sample
  When program run with camera RTSP stream at resolution 1280x720, first video picture is good :
  First good at res.1280x720
  but then not good :
  not good at res.1280x720
  Its seem to be my FFmpeg function call to avcodec_decode_video2 can’t fully decode certain packet for some reasons.