Recherche avancée

Sur d’autres sites (9569)

• libav C++ sw_scale returns variable byte size of BGR24 buffer

  6 mars 2021, par igor

  I'm trying to convert my local /dev/video0 cam to BGR24 format. It works if I resize the image (although not 100% of the time, more like 85% of the time), but I'd like to keep the same size as input video.

  


  I initialize BGR image like so including the sws context :

  


  AVPixelFormat outputPixFormat = AV_PIX_FMT_BGR24;

    AVFrame* pFrameBGR = av_frame_alloc();
    pFrameBGR->width = decoder->video_codec_context->width;
    pFrameBGR->height = decoder->video_codec_context->height;
    pFrameBGR->format = outputPixFormat;

    int alloRet = av_image_alloc(pFrameBGR->data, pFrameBGR->linesize, decoder->video_codec_context->width, decoder->video_codec_context->height, outputPixFormat, 1);
    if (alloRet < 0) {
        logging("failed to allocate image");
        return -1;
    }

    struct SwsContext *sws_ctx = NULL;

    sws_ctx = sws_getContext(decoder->video_codec_context->width,
    decoder->video_codec_context->height,
    decoder->video_codec_context->pix_fmt,
    decoder->video_codec_context->width,
    decoder->video_codec_context->height,
    outputPixFormat,
    SWS_DIRECT_BGR,
    0,
    0,
    0
    );


  


  This is the portion of my decoding loop :

  


  int response = avcodec_send_packet(pCodecContext, pPacket);
    if (response < 0) {
        logging("Error while sending a packet to decoder: %s", av_err2str(response));
        return response;
    }

    while (response >= 0) {
        response = avcodec_receive_frame(pCodecContext, pFrame);
         if (response == AVERROR(EAGAIN) || response == AVERROR_EOF) {
            break;
        } else if (response < 0) {
            logging("Error while receiving a frame from the decoder: %s", av_err2str(response));
            return response;
        }
        if (response >= 0) {

            sws_scale(sws_ctx,  (uint8_t const * const *)pFrame->data, pFrame->linesize, 0, pCodecContext->height, pFrameBGR->data, pFrameBGR->linesize);



  


  THe question is how to copy the plane of AVFrame into a buffer :

  


    size_t rgb_size = av_image_get_buffer_size(AV_PIX_FMT_BGR24, bgrFrame->width, bgrFrame->height, 1);

    uint8_t *dst_data;
    dst_data = (uint8_t *)(av_malloc(rgb_size));

    av_image_copy_to_buffer(dst_data, rgb_size, (const uint8_t* const *)bgrFrame->data, bgrFrame->linesize, AV_PIX_FMT_BGR24, bgrFrame->width, bgrFrame->height, 1);



  


  If I try to save to file the BGR image is correctly copied :

  


  char filebuf[256];
snprintf(filebuf, sizeof filebuf, "%s%d%s", "out_", pPacket->dts, ".rgb");
std::FILE *output=fopen(filebuf,"wb+");  

fwrite(bgrFrame->data[0],(pFrame->width)*(pFrame->height)*3,1,output);  
std::fclose(output);


  


  So it looks like my copy to buffer function is faulty, but I can figure out what's wrong with it :

  


    uint8_t *dst_data;
    dst_data = (uint8_t *)(av_malloc(rgb_size));

    av_image_copy_to_buffer(dst_data, rgb_size, (const uint8_t* const *)bgrFrame->data, bgrFrame->linesize, AV_PIX_FMT_BGR24, bgrFrame->width, bgrFrame->height, 1);


  


• problem with creating a robot voice with flutterFFmpeg

  28 février 2021, par Sadiq Samaila

  so here is the code I ran

  


                        String commandToExecute =
                      "-i $aud -filter_complex 'afftfilt=real='hypot(re,im)*sin(0)':imag='hypot(re,im)*cos(0)':win_size=512:overlap=0.75' " +
                          outputPath;
   
                  _flutterFFmpeg.execute(commandToExecute).then((rc) => print(
                      "FFmpeg process exited with rc $rc $appDocumentDir"));


  


  and I got this error

  


  E/mobile-ffmpeg( 5493): [AVFilterGraph @ 0x7ce619159800] No such filter: 'im)*sin(0):imag'
E/mobile-ffmpeg( 5493): Error initializing complex filters.
E/mobile-ffmpeg( 5493): Invalid argument
D/flutter-ffmpeg( 5493): FFmpeg exited with rc: 1


  


  I tried changing all single quote(') to double quote(") in the command but it did not work.

  


• Revision 9068bce4e7 : Put iterative motion search under speed control Enable iterative motion search

  3 juin 2013, par Jingning Han

  Changed Paths :
   Modify /vp9/encoder/vp9_rdopt.c

  
  
  Put iterative motion search under speed control

  Enable iterative motion search for compound inter-inter prediction
  of block sizes 4x4/4x8/8x4 only when best coding quality is selected.
  The iterative motion search provides about 0.1% gains for derf and
  stdhd at this point, at the expense of longer runtime.

  Change-Id : Idc03e7f827e51f1bb8d269bc3752ee297a6bbfe5