Recherche avancée

Sur d’autres sites (4312)

• FFmpeg4 android Applying Overlay (watermark) taking too long

  13 janvier 2019, par Web Developer

  I’m using ffmpeg4 on android (Java) to apply overlay/watermark on the bottom left corner of the recorded video that is saved in the SD CARD.

  The video recording has maximum duration of 59 seconds. The problem that I’m facing is that I’m applying water through this command of ffmpeg4

  String[] complexCommand1 = new String[]"ffmpeg", "-y", "-i",
  "input.mp4", "-strict", "experimental", "-c:v", "libx264", "-preset",
  "ultrafast", "-vf", "movie=/water.png [watermark] ; [in][watermark] overlay=main_w-overlay_w-", "10:10", " [out]", "-s", "320x480", "-aspect",
  "2:3", "-r", "30", "-b:v", "500k",
  "output.mp4"  ;

  I’ve reduced the video resolution to 320x480 after the conversion and 480p to recorded video that is provided as input to this command but still the processing time is around 40 seconds.

  Can someone provide me workaround this to reduce this processing time.

  or what should I do to reduce this processing time. I’ve researched a lot on this and found some GitHub issues link and tried their solutions but none of them seemed to work for me.

• FPS from RTSP stream info does not match actual framerate

  17 mai 2021, par Krapow

  I have a 25FPS RTSP stream coming from an IP-camera. I can successfully display the video stream. But when analyzing the stream with ffmpeg (ffprobe actually), I observe fewer frames per second rate :

  


  $ ffprobe -rtsp_transport tcp -i rtsp://camera_ip:554/stream -select_streams v:0 -show_frames -show_entries frame=coded_picture_number,pkt_pts_time -of csv=p=0
Stream #0:0: Video: h264 (Main), yuvj420p(pc, bt709, progressive), 640x480, 25 fps, 25 tbr, 90k tbn, 50 tbc
0.400000,0
0.080000,1
0.120000,2
0.200000,3
0.240000,4
0.320000,5
0.360000,6
0.440000,7
0.480000,8
0.560000,9
0.600000,10
0.680000,11
0.720000,12
0.800000,13
0.840000,14
0.920000,15
0.960000,16
1.040000,17
1.080000,18
1.160000,19
1.200000,20
1.280000,21
1.320000,22
1.400000,23
1.440000,24
1.520000,25
1.560000,26
1.640000,27
1.680000,28
1.760000,29
1.800000,30
1.880000,31
1.920000,32
2.000000,33


  


  We can clearly see the 80ms gap between some of the frames, resulting in a 16fps stream.

  


  I have observed the same framerate issue with GStreamer (printing information in the rtpjitterbuffer indicates the frame gap is sometimes 80ms and sometimes 40ms). But the weird thing is, I encountered the same issue with an HDMI-RJ45 decoder, and I doubt the same issue comes from 2 different devices.
I didn't get much more informations using -loglevel debug or trace.
Does anybody have an idea about what is going wrong in the stream ?

  


  (I used ffprobe 4.2.3 and the last "2021-05-09-git-8649f5dca6-full_build-www.gyan.dev" with the same results, and GStreamer 1.16.2 with a pipeline like "urisourcebin ! h264depay ! h264parse ! fakesink")

  


  EDIT : The camera skipping of frames was caused by the activation of a third stream in the options. I find it really weird that it skips exactly the same frames every seconds. However, I still haven't found the cause of the downrate on my RTSP encoder.
Anyway, this was actually hardware related and not software related.

  


• Does this code contain a potential memory leak ?

  26 juillet 2017, par Johnnylin

  Here is the code :

  cv::Mat YV12ToBRG24_FFmpeg(unsigned char* pYUV, int width,int height)
  
  {
  
      if (width < 1 || height < 1 || pYUV == nullptr){
  
          return cv::Mat();
  
      }
  
  
  
      cv::Mat result(height,width,CV_8UC3, cv::Scalar::all(0));
  
      uchar* pBGR24 = result.data;
  
  
  
      AVFrame* pFrameYUV = av_frame_alloc();
  
      pFrameYUV->width = width;
  
      pFrameYUV->height = height;
  
      pFrameYUV->format = AV_PIX_FMT_YUV420P;
  
      av_image_fill_arrays(pFrameYUV->data, pFrameYUV->linesize, pYUV, AV_PIX_FMT_YUV420P, width, height, 1);
  
  
  
      //U,V exchange
  
      uint8_t * ptmp=pFrameYUV->data[1];
  
      pFrameYUV->data[1]=pFrameYUV->data[2];
  
      pFrameYUV->data[2]=ptmp;
  
  
  
      AVFrame* pFrameBGR = av_frame_alloc();
  
      pFrameBGR->width = width;
  
      pFrameBGR->height = height;
  
      pFrameBGR->format = AV_PIX_FMT_BGR24;
  
      av_image_fill_arrays(pFrameBGR->data, pFrameBGR->linesize, pBGR24, AV_PIX_FMT_BGR24, width, height, 1);
  
  
  
      struct SwsContext* imgCtx = nullptr;
  
      imgCtx = sws_getContext(width,height,AV_PIX_FMT_YUV420P,width,height,AV_PIX_FMT_BGR24,SWS_BILINEAR,0,0,0);
  
      if (imgCtx != nullptr){
  
          sws_scale(imgCtx,pFrameYUV->data,pFrameYUV->linesize,0,height,pFrameBGR->data,pFrameBGR->linesize);
  
          sws_freeContext(imgCtx);
  
          imgCtx = nullptr;
  
          ptmp = nullptr;
  
          pBGR24 = nullptr;
  
          av_frame_free(&pFrameYUV);
  
          av_frame_free(&pFrameBGR);
  
          return result;
  
      }
  
      else{
  
          sws_freeContext(imgCtx);
  
          imgCtx = nullptr;
  
          ptmp = nullptr;
  
          pBGR24 = nullptr;
  
          av_frame_free(&pFrameYUV);
  
          av_frame_free(&pFrameBGR);
  
          return cv::Mat();
  
      }
  
  }

  This function is called every 40 ms (25 fps) and I saw a significant memory increase after several days(like 12GB). But if I ran this code for hours, the memory leak problem would not be obvious enough to be observed.

  Can anyone help ?
  Thanks.