Recherche avancée

Sur d’autres sites (6133)

• FFmpeg memory leak

  10 septembre 2015, par Spamdark

  I have developed just a simple library modifing a library that I found on the internet.

  What scares me, is that, when I play an avi, it plays and free the memory when the video ends, but when I play the video, it’s like a memory leak ! It grows to 138mb although the video has ended and the FreeAll method (A function that deletes the context, etc...) has been called.

  Here is the code of the method that is causing the memory leak :

  int VideoGL::NextVideoFrame(){
  
  int frameDone = 0;
  
  int result = 0;
  
  double pts = 0;
  
  
  
  if(!this->ended){
  
  
  
  if (!_started) return 0;
  
  AVPacket* packet;
  
  
  
  // Get the number of milliseconds passed and see if we should display a new frame
  
  int64_t msPassed = (1000 * (clock() - _baseTime)) / CLOCKS_PER_SEC;
  
  if (msPassed >= _currentPts)
  
  {
  
      // If this is not the current frame, copy it to the buffer
  
      if (_currentFramePts != _currentPts){
  
          _currentFramePts = _currentPts;
  
          memcpy(buffer_a,buffer, 3 * _codec_context_video->width * _codec_context_video->height);
  
          result = 1;
  
      }
  
  
  
      // Try to load a new frame from the video packet queue
  
      bool goodop=false;
  
      AVFrame *_n_frame = avcodec_alloc_frame();
  
      while (!frameDone && (packet = this->DEQUEUE(VIDEO)) != NULL)
  
      {
  
          if (packet == (AVPacket*)-1) return -1;
  
  
  
          goodop=true;
  
  
  
          _s_pts = packet->pts;
  
          avcodec_decode_video2(_codec_context_video, _n_frame, &frameDone, packet);
  
          av_free_packet(packet);
  
  
  
          if (packet->dts == AV_NOPTS_VALUE)
  
          {
  
              if (_n_frame->opaque && *(uint64_t*)_n_frame->opaque != AV_NOPTS_VALUE) pts = (double) *(uint64_t*)_n_frame->opaque;
  
              else pts = 0;
  
          }
  
          else pts = (double) packet->dts;
  
  
  
          pts *= av_q2d(_codec_context_video->time_base);
  
  
  
      }
  
  
  
      if (frameDone)
  
      {
  
          // if a frame was loaded scale it to the current texture frame buffer, but also set the pts so that it won't be copied to the texture until it's time
  
          sws_scale(sws_ctx,_n_frame->data, _n_frame->linesize, 0, _codec_context_video->height, _rgb_frame->data, _rgb_frame->linesize);
  
  
  
  
  
          double nts = 1.0/av_q2d(_codec_context_video->time_base);
  
          _currentPts = (uint64_t) (pts*nts);
  
  
  
      }
  
  
  
      avcodec_free_frame(&_n_frame);
  
      av_free(_n_frame);
  
  
  
      if(!goodop){
  
          ended=true;
  
      }
  
  }
  
  }
  
  
  
  return result;
  
  }

  I’ll be waiting for answers, thanks.

• Android how to increase ffmpeg mp4 perfromance ?

  4 janvier 2013, par testCoder

  I have detected that function avcodec_decode_audio3 works slow with mp4 format, here my code cycle for decoding audio :

  while (av_read_frame(av_format_context, &packet) >= 0 && is_play == 1) {
  
          if (av_codec_context->codec_type == AVMEDIA_TYPE_AUDIO
  
                  && is_play == 1) {
  
              int out_size = AVCODEC_MAX_AUDIO_FRAME_SIZE;
  
              int size = packet.size;
  
              int n;
  
              int dataLength = size;
  
              int decoded = 0;
  
              while (size > 0) {
  
  
  
                  //start measure time
  
                  gettimeofday(&tvBegin, NULL);
  
  
  
                  int len = avcodec_decode_audio3(av_codec_context,
  
                          (int16_t *) pAudioBuffer, &out_size, &packet);
  
  
  
                  //stop measure time
  
                  gettimeofday(&tvEnd, NULL);
  
                  timeval_subtract(&tvDiff, &tvEnd, &tvBegin);
  
  
  
                  LOGI("%d", tvDiff.tv_usec / 1000);
  
                  LOGI("len='%d'", len);
  
                  LOGI("out_size='%d'", out_size);
  
  
  
                  if (len < 0) {
  
                      break;
  
                      return 1;
  
                  }
  
                  if (out_size > 0) {
  
  
  
  
  
  
  
                      jbyte *bytes = (*env)->GetByteArrayElements(env, array,
  
                              NULL);
  
                      memcpy(bytes, (int16_t *) pAudioBuffer, out_size);
  
                      (*env)->ReleaseByteArrayElements(env, array, bytes, 0);
  
                      (*env)->CallVoidMethod(env, obj, play, array, out_size,
  
                              is_play);
  
  
  
                  }
  
                  size -= len;
  
              }
  
  
  
          }
  
          if (packet.data)
  
              av_free_packet(&packet);
  
  
  
      }

  But with other formats like flac and mp3 it works fine. avcodec_decode_audio3 take about 1-2 milisecounds for decoding mp3 frame with out_size = 4608 but with the same frame size in mp4 decoding take about 6-7 millisecounds. I got my build script from here.

  Does it normal behavior ? Is any way to increase performance of decoding mp4 ?

• When using ffmpeg to create mp4 video file from batch of images the whole process is very slow how can i make it faster ?

  29 juin 2015, par Brubaker Haim

  The whole process is slow and also in the end the video file when playing it the frames moving very slow.

  ffmpeg -framerate 1/5 -i screenshot%06d.jpg -c:v libx264 -r 30 -p
  
  ix_fmt yuv420p out2.mp4

  Is that mean 1 frames each 5 seconds ?
  So if i will make 5/1 it will be 5 frames in a second ?
  What should be the best result ?

  And the second problem is that for testing i have 70 images but in the original i have over 1000 images is there any way to make all this process faster ?