Recherche avancée

Sur d’autres sites (7012)

• How to create video using avcodec from jpeg images of type OpenCV::Mat ?

  23 juillet 2015, par theateist

  I have colored jpeg images of OpenCV::Mat type and I create from them video using avcodec. The video that I get is upside-down, black & white and each row of each frame is shifted and I got diagonal line. What could be the reason for such output ?
  Follow this link to watch the video I get using avcodec.
  I’m using acpicture_fill function to create avFrame from cv::Mat frame !

  P.S.
  Each cv::Mat cvFrame has width=810, height=610, step=2432
  I noticed that avFrame (that is filled by acpicture_fill) has linesize[0]=2430
  I tried manually setting avFrame->linesizep0]=2432 and not 2430 but it still didn’t helped.

  ======== CODE =========================================================

  AVCodec *encoder = avcodec_find_encoder(AV_CODEC_ID_H264);
  
  AVStream *outStream = avformat_new_stream(outContainer, encoder);
  
  avcodec_get_context_defaults3(outStream->codec, encoder);
  
  
  
  outStream->codec->pix_fmt = AV_PIX_FMT_YUV420P;
  
  outStream->codec->width = 810;
  
  outStream->codec->height = 610;
  
  //...
  
  
  
  SwsContext *swsCtx = sws_getContext(outStream->codec->width, outStream->codec->height, PIX_FMT_RGB24,
  
                                      outStream->codec->width, outStream->codec->height,  outStream->codec->pix_fmt, SWS_BICUBIC, NULL, NULL, NULL);
  
  
  
  for (uint i=0; i < frameNums; i++)
  
  {
  
      // get frame at location I using OpenCV
  
      cv::Mat cvFrame;
  
      myReader.getFrame(cvFrame, i); 
  
      cv::Size frameSize = cvFrame.size();    
  
      //Each cv::Mat cvFrame has  width=810, height=610, step=2432
  
  
  
  
  
  1.  // create AVPicture from cv::Mat frame
  
  2.  avpicture_fill((AVPicture*)avFrame, cvFrame.data, PIX_FMT_RGB24, outStream->codec->width, outStream->codec->height);
  
  3avFrame->width = frameSize.width;
  
  4.  avFrame->height = frameSize.height;
  
  
  
      // rescale to outStream format
  
      sws_scale(swsCtx, avFrame->data, avFrame->linesize, 0, outStream->codec->height, avFrameRescaledFrame->data, avFrameRescaledFrame ->linesize);
  
  encoderRescaledFrame->pts=i;
  
  avFrameRescaledFrame->width = frameSize.width;
  
      avFrameRescaledFrame->height = frameSize.height;
  
  
  
  av_init_packet(&avEncodedPacket);
  
      avEncodedPacket.data = NULL;
  
      avEncodedPacket.size = 0;
  
  
  
      // encode rescaled frame
  
      if(avcodec_encode_video2(outStream->codec, &avEncodedPacket, avFrameRescaledFrame, &got_frame) < 0) exit(1);
  
      if(got_frame)
  
      {
  
          if (avEncodedPacket.pts != AV_NOPTS_VALUE)
  
              avEncodedPacket.pts =  av_rescale_q(avEncodedPacket.pts, outStream->codec->time_base, outStream->time_base);
  
          if (avEncodedPacket.dts != AV_NOPTS_VALUE)
  
              avEncodedPacket.dts = av_rescale_q(avEncodedPacket.dts, outStream->codec->time_base, outStream->time_base);
  
  
  
          // outContainer is "mp4"
  
          av_write_frame(outContainer, & avEncodedPacket);
  
  
  
          av_free_packet(&encodedPacket);
  
      }
  
  }

  UPDATED

  As @Alex suggested I changed the lines 1-4 with the code below

  int width = frameSize.width, height = frameSize.height; 
  
  avpicture_alloc((AVPicture*)avFrame, AV_PIX_FMT_RGB24, outStream->codec->width, outStream->codec->height);
  
  for (int h = 0; h < height; h++)
  
  {
  
       memcpy(&(avFrame->data[0][h*avFrame->linesize[0]]), &(cvFrame.data[h*cvFrame.step]), width*3);
  
  }

  The video (here) I get now is almost perfect. It’s NOT upside-down, NOT black & white, BUT it seems that one of the RGB components is missing. Every brown/red colors became blue (in original images it should be vice-verse).
  What could be the problem ? Could rescaling(sws_scale) to AV_PIX_FMT_YUV420P format causes this ?

• Overlay two sequences of png's and turn them into a movie

  12 mai 2014, par Lau Llobet

  I’ve found how to turn a png sequence into a movie and I’ve found how to overlay two movies using transparency but I don’t know how to do both things at once (using png’s transparency).

  The bottom layer of png’s is smaller than the top one and needs to be stretched to a certain resolution and also have a padding to be centered.

  The output doesn’t have to have alpha (black for alpha is ok). I’m a bit confused by the abundance of filter options.

  For the moment I’ve found :

  ./ffmpeg -i ./seq1/%d.bmp -vf "movie=./%d.png [a]; [in][a] overlay=0:366" combined.m2v

  It works, now I’ve got to find the padding and resize things

• Revision 734c5ffa2c : Apply constrained partition search range to non-RD mode decision This commit en

  23 avril 2014, par Jingning Han

  Changed Paths :
   Modify /vp9/encoder/vp9_encodeframe.c

  
  
  Apply constrained partition search range to non-RD mode decision

  This commit enables a chessboard pattern for partition search. All
  the black blocks run regular partition search ranging from 8x8 to
  32x32. The rest white blocks take the nearby blocks’ information
  to adaptively decide the effective search range.

  The compression performance for rtc set at speed -5 is down by 1.5%.
  For pedestrian 1080p at speed -5, the runtime goes from 41594 ms to
  39697 ms, i.e., about 5% faster.

  Change-Id : Ia4b96e237abfaada487c743bca08fe1afd298685