Recherche avancée

Sur d’autres sites (5045)

• Webrtc streaming issue with Wowza and FFMPEG

  26 juin 2018, par Diego T

  I am trying to stream video and audio from a Camera in a browser using Webrtc and Wowza Media Server (4.7.3 version).

  The camera stream (h264/aac) is first of all transcoded by using FFMPEG (version N-89681-g2477bfe built with gcc 4.8.5, last available version on ffmpeg website) in VP8/OPUS and then pushed to the Wowza Server.
  By using the small Wowza webpage I ask for the Wowza stream to be displayed in the browser (Chrome Version 66.0.3336.5 Build officiel canary 32 bits).

  FFMPEG used command :

  ffmpeg -rtsp_transport tcp -i rtsp:// -vcodec libvpx -vb 600000 -crf 10 -qmin 0 -qmax 50 -acodec libopus -ab 32000 -ar 48000 -ac 2 -f rtsp rtsp://://test

  When I click on Play stream I have a very bad quality video and audio (jerky video and very bad audio).

  If I use this FFMPEG command :

  ffmpeg -rtsp_transport tcp -i rtsp:// -vcodec libvpx -vb 600000 -crf 10 -qmin 0 -qmax 50 -acodec copy -f rtsp rtsp://://test

  I will have a good video (flowing, smooth) but no audio (the camera micro is ON).

  If libopus is the problem (as this test first shows), I tried libvorbis but with Chrome console I have this error "Failed to set remote offer sdp : Session error code : ERROR_CONTENT". Weird, cause libvorbis is one of the available codecs for Webrtc.

  Is someone experiencing the same issue ? Did someone experience the same issue ?

  Thanks in advance.

• QOpenGLWidget video rendering perfomance in multiple processes

  24 novembre 2018, par MasterAler

  My problem may seem vague without code, but it actually isn’t.

  So, there I’ve got an almost properly-working widget, which renders video frames.

  Qt 5.10 and QOpenGLWidget subclassing worked fine, I didn’t make any sophisticated optimizations — there are two textures and a couple of shaders, converting YUV pixel format to RGB — glTexImage2D() + shaders, no buffers.

  Video frames are obtained from FFMPEG, it shows great performance due to hardware acceleration... when there is only one video window.

  The piece of software is a "video wall" — multiple independent video windows on the same screen. Of course, multi-threading would be the preferred solution, but legacy holds for now, I can’t change it.

  So, 1 window with Full HD video consumes 2% CPU & 8-10% GPU regardless of the size of the window. But 7-10 similar windows, launched from the same executable at the same time consume almost all the CPU. My math says that 2 x 8 != 100...

  My best guesses are :

  • This is a ffmpeg decoder issue, hardware acceleration still is not magic, some hardware pipeline stalls
  • 7-8-9 independent OpenGL contexts cost a lot more than 1 cost x N
  • I’m not using PUBO or some other complex techniques to improve OpenGL rendering. It still explains nothing, but at least it is a guess

  The behavior is the same on Ubuntu, where decoding uses different codec (I mean that using GPU accelerated or CPU accelerated codecs makes no difference !), so, it makes more probable that I’m missing something about OpenGL... or not, because launching 6-7 Qt examples with dynamic textures shows normal growth of CPU usages — it is approximately a sum for the number of windows.

  Anyway, it becomes quite tricky for me to profile the case, so I hope someone could have been solving the similar problem before and could share his experience with me. I’d be appreciated for any ideas, how to deal with the described riddle.

  I can add any pieces of code if that helps.

• Interact with ffmpeg from a .NET program - Write Input

  7 mai 2015, par Shimmy

  In reference to this question, as you can see I managed to run and receive data from the program.

  However I didn’t manage to submit data to it, for instance, while converting a file, pressing q immediately stop conversion and stops the program.
  I need my application to support stopping the process as well, and I think this should be done by passing this parameter to the ffmpeg app, since I want it to take care of all uncollected resource or whatever dust it would leave behind if I would just go and use process.Kill()

  Here is what I’ve tried :

  static int lineCount = 0;
  
  static bool flag;
  
  static void process_ErrorDataReceived(object sender, DataReceivedEventArgs e)
  
  {
  
    Console.WriteLine("Error ({1:m:s:fff}: {0})", lineCount++,
  
        DateTime.Now);
  
  
  
    if (e.Data != null && string.Equals(e.Data,"Press [q] to stop, [?] for help")) 
  
      flag = true;
  
  
  
    if (flag)
  
    {
  
      flag = false;
  
      Console.WriteLine("Stopping ({0:m:s:fff})...", DateTime.Now);
  
      process.CancelErrorRead();
  
      process.CancelOutputRead();
  
      process.StandardInput.WriteLine("q");
  
    }   
  
  
  
    Console.WriteLine(e.Data);
  
    Console.WriteLine();
  
  }

  But it doesn’t do anything, seems that once the conversion has been requested, I have no control on it any more, I can only receive output from it. Running it as stand alone does allow me interaction of course.

  What am I missing here, is it a different trick in submitting the output or the code in previous answer is wrong, or I should have chosen a different approach ?

  For your attention, RedirectStandardInput is on.

  NOTE : as you can see in the answer of my previous question, ffmpeg interacts differently, I think the one who knows the answer will be (maybe I’m wrong) someone with experience in ffmpeg.