Recherche avancée

Sur d’autres sites (6578)

• FFmpeg can not open video file after adding the GLsurfaceView to render frames

  4 avril 2016, par Kyle Lo

  The source code works perfectly without any modification.

  I successfully use the below function to play the specified video.

  playview.openVideoFile("/sdcard/Test/mv.mp4");

  And for the research purpose I need to display the frame by using OpenGL ES. So I remove the original method below.

  ANativeWindow* window = ANativeWindow_fromSurface(env, javaSurface);
  
  
  
  ANativeWindow_Buffer buffer;
  
  if (ANativeWindow_lock(window, &buffer, NULL) == 0) {
  
    memcpy(buffer.bits, pixels,  w * h * 2);
  
    ANativeWindow_unlockAndPost(window);
  
  }
  
  
  
  ANativeWindow_release(window);

  And I add FrameRenderer class into my project

  public class FrameRenderer implements GLSurfaceView.Renderer {
  
  
  
      public long time = 0;
  
      public short framerate = 0;
  
      public long fpsTime = 0;
  
      public long frameTime = 0;
  
      public float avgFPS = 0;
  
      private PlayNative mNative = null;
  
  
  
      @Override
  
      public void onSurfaceCreated(GL10 gl, EGLConfig config) {/*do nothing*/}
  
  
  
      @Override
  
      public void onSurfaceChanged(GL10 gl, int width, int height) {
  
  
  
      }
  
  
  
      @Override
  
      public void onDrawFrame(GL10 gl) {
  
          mNative.render();
  
      }

  In the native side I create a corresponding method in VideoPlay.cpp And I only use glClearColorto test if the OpenGL function works or not.

  void VideoPlay::render() {
  
      glClearColor(1.0f, 0.0f, 0.0f, 1.0f);
  
      glClear(GL_COLOR_BUFFER_BIT);
  
  }

  And the onCreate is as below.

  protected void onCreate(Bundle savedInstanceState) {
  
          // TODO Auto-generated method stub
  
          super.onCreate(savedInstanceState);
  
          setContentView(R.layout.main_layout);
  
  
  
          playview = new PlayView(this);
  
  
  
          playview.openVideoFile("/sdcard/test_tt_racing.mp4");
  
          //playview.openVideoFile("/sdcard/big_buck_bunny.mp4");
  
  
  
          GLSurfaceView surface = (GLSurfaceView)findViewById(R.id.surfaceviewclass);
  
          surface.setRenderer(new FrameRenderer());
  
          ...

  Then test it on the mobile, the screen becomes red which means the GLSurfaceView and OpenGL works fine.

  But after I press the play bottom, whole the app stucked. And Show in the
  Log

  My question is why I can open the video whose path is totally the same with the previous one, just after I added the GLsurface renderer and how can I fix it ?

• FFMPEG - FFV1 frame encoding crashes on cleaning up

  21 mai 2016, par Yan

  I’m trying to implement a frame encoding functionality using the ffmpeg c-api. I am receiving frames from a camera which are in the Gray16le format. I want to convert encode them using the ffv1 encoder and copy the resulting frame into the variable "data". This is the code that I got so far. It seems to be working in a sense that it doesn’t crash until the part where I am freeing up my variables.

  /* Video compression variables///////////////////////////////////*/
  
  struct timeval stop, start;
  
  
  
  AVCodec *codec;
  
  AVCodecContext *context= NULL;
  
  const AVPixFmtDescriptor *avPixDesc = NULL; // used to get bits per pixel
  
  int ret, got_output;
  
  int bufferSize = 0; // Size of encoded image frame in bytes
  
  //uint8_t* outBuffer;
  
  AVFrame *inFrame; //
  
  AVPacket pkt;
  
  Data* data;
  
  /* Video compression ///////////////////////////////////*/
  
  
  
  Frame* frame;
  
  /////////////////////////////////////////////////////////////////////////
  
  // start frame compression - current codec is ffv1
  
  //////////////////////////////////////////////////////////////////////////
  
  
  
  gettimeofday(&start, NULL); // get current time
  
  avcodec_register_all(); // register all the codecs
  
  codec = avcodec_find_encoder(AV_CODEC_ID_FFV1); // find the ffv1 encoder
  
  if (!codec) {
  
      fprintf(stderr, "Codec not found\n");
  
      exit(1);
  
  }
  
  
  
  context = avcodec_alloc_context3(codec);
  
  if (!context) {
  
      fprintf(stderr, "Could not allocate video codec context\n");
  
      exit(1);
  
  }
  
  
  
  frame = getFrame(); // get frame so we can set context params
  
  
  
  /* put sample parameters */
  
  context->bit_rate = 400000; // from example, half might also work
  
  /* resolution must be a multiple of two */
  
  context->width = frame->size[0];
  
  context->height = frame->size[1];
  
  /* frames per second */
  
  context->time_base = (AVRational){1,22}; // 22 fps
  
  
  
  context->gop_size = 1; // typical for ffv1 codec
  
  context->max_b_frames = 1; // set to 1 for now, the higher the b-frames count, the higher the needed ressources
  
  context->pix_fmt = AV_PIX_FMT_GRAY16LE ; // same as source, Y , 16bpp, little-endian, 12 of the 16 pixels are used
  
  
  
  /* open it */
  
  if (avcodec_open2(context, codec, NULL) < 0) {
  
          fprintf(stderr, "Could not open codec\n");
  
          exit(1);
  
  }
  
  
  
  inFrame = av_frame_alloc();
  
  if(!inFrame)
  
  {
  
      printf("Could not allocate video frame\n! Exciting..");
  
      exit(1);
  
  }
  
  
  
  // allocate image in inFrame
  
  ret = av_image_alloc(inFrame->data, inFrame->linesize, context->width, context->height, context->pix_fmt, 16);
  
  if(ret<0)
  
  {
  
      printf("Error allocating image of inFrame! Exiting..\n");
  
      exit(1);
  
  }
  
  
  
  // copy data of frame of type Frame* into frame of type AVFrame* so we can use ffmpeg to encode it
  
  int picFill = avpicture_fill((AVPicture*)inFrame, (uint8_t*)frame->image, context->pix_fmt, context->width, context->height);
  
  
  
  if(picFill < 0)
  
  {
  
      printf("Error filling inFrame with frame->image! Exiting..\n");
  
      exit(1);
  
  }
  
  else
  
  {
  
      printf("Successfully filled inFrame with frame->image..\n");
  
      printf("Size of bytes filled:  %d", picFill);
  
  }
  
  
  
  inFrame->width = context->width;
  
  inFrame->height = context->height;
  
  inFrame->format = context->pix_fmt;
  
  
  
  if(frame->image[0] == NULL)
  
  {
  
          printf("Error! frame->image[0] == NULL.. Exiting..\n");
  
          exit(1);
  
  }
  
  
  
  fflush(stdout);
  
  int i=0;
  
  
  
  // start encoding
  
  while(!got_output) // while we didn't get a complete package
  
  {
  
      /* Start encoding the given frame */
  
      av_init_packet(&pkt);
  
      pkt.data = NULL;    // packet data will be allocated by the encoder
  
      pkt.size = 0;
  
  
  
      i++;
  
  
  
      /* encode the image */
  
      ret = avcodec_encode_video2(context, &pkt, inFrame, &got_output);
  
      if (ret < 0) {
  
              fprintf(stderr, "Error encoding frame\n");
  
              exit(1);
  
      }
  
  
  
      inFrame->pts = i;
  
  
  
      if(got_output)
  
      {
  
          printf("Got a valid package after %d frames..\n", i);
  
          // encoding of frame done, adapt "data"-field accordingly
  
          avPixDesc = av_pix_fmt_desc_get(context->pix_fmt); // Get pixelFormat descriptor
  
          bufferSize = av_image_get_buffer_size(context->pix_fmt, inFrame->width, inFrame->height,16);
  
          if(bufferSize <= 0)
  
          {
  
              printf("Error! Buffersize of encoded frame is <= 0, exciting...\n");
  
          }
  
          else
  
          {
  
              printf("Buffersize determined to be %d\n", bufferSize);
  
          }
  
  
  
          data->m_size[0] = inFrame->width;
  
          data->m_size[1] = inFrame->height;
  
          data->m_bytesPerPixel = av_get_bits_per_pixel(avPixDesc)/8;
  
  
  
          if (0 != av_get_bits_per_pixel(avPixDesc) % 8)
  
                  data->m_bytesPerPixel += 1;
  
  
  
          printf("Buffersize is: %d, should be %d\n", bufferSize, inFrame->width * inFrame->height * data->m_bytesPerPixel);
  
          data->m_image = malloc(bufferSize);
  
          printf("copying data into final variable...\n");
  
  
  
          memcpy(data->m_image, pkt.data, bufferSize); // copy data from ffmpeg frame
  
          printf("copying of data done\n");
  
  
  
          printf("Unrefing packet..\n");
  
          av_packet_unref(&pkt);
  
          printf("Unrefing packet done..\n");
  
      }
  
      else
  
      {
  
          printf("Didnt get package, so we get and encode next frame..\n");
  
          frame = getFrame(); // get next frame             
  
  
  
          picFill = avpicture_fill((AVPicture*)inFrame, (uint8_t*)frame->image, context->pix_fmt, context->width, context->height);
  
          if(!picFill)
  
          {
  
              printf("Error filling frame with data!!..\n");
  
              exit(1);
  
          }
  
          else
  
          {
  
              printf("Size required to store received frame in AVFrame in bytes: %d", picFill);
  
          }
  
      }
  
  }
  
  
  
  printf("\nDone with encoding.. cleaning up..\n");
  
  printf("Closing context...\n");
  
  avcodec_close(context);
  
  printf("Closing context done...\n");
  
  printf("Freeing context...\n");
  
  av_free(context);
  
  printf("Freeing context done...\n");
  
  if(inFrame->data[0] != NULL)
  
  {
  
      printf("avfreep() pointer to FFMPEG frame data...\n");
  
      av_freep(&inFrame->data[0]);
  
      printf("Freeing pointer to FFMPEG frame data done...\n");
  
  }
  
  else
  
  {
  
      printf("infRame->data[0] was not deleted because it was NULL\n");
  
  }
  
  
  
  printf("Freeing frame...\n");
  
  av_frame_free(&inFrame);
  
  printf("Freeing inFrame done...\n");
  
  printf("Compression of frame done...\n");
  
  gettimeofday(&stop, NULL);
  
  printf("took %lu ms\n", (stop.tv_usec - start.tv_usec) / 1000);

  This is the output that I am getting when I run the program :

  [ffv1 @ 0x75101970] bits_per_raw_sample > 8, forcing range coder
  
  Successfully filled inFrame with frame->image..
  
  Size of bytes filled:  1377792Got a valid package after 1 frames..
  
  Buffersize determined to be 1377792
  
  Buffersize is: 1377792, should be 1377792
  
  copying data into final variable...
  
  copying of data done
  
  Unrefing packet..
  
  Unrefing packet done..
  
  
  
  Done with encoding.. cleaning up..
  
  Closing context...
  
  Closing context done...
  
  Freeing context...
  
  Freeing context done...
  
  avfreep() pointer to FFMPEG frame data...
  
  *** Error in `./encoding': free(): invalid pointer: 0x74a66428 ***
  
  Aborted

  The error seems to occur when calling the av_freep() function. If you could point me in the right direction, it would be greatly appreciated ! This is my first time working with the ffmpeg api and I feel that I am not so close to my goal, though I spent quite some time looking for the error already..

  Best regards !

• Proper reading of MP3 file disrupted by ID3 tags

  3 septembre 2016, par PookyFan

  My semestral project is due this Thursday and I have major problem with reading MP3 file (the project is about sound analysis, don’t ask my what exactly is it about and why I’m doing it so late).

  First, I read first 10 bytes to check for ID3 tags. If they’re present, I’ll just skip to the first MP3 header - or at least that’s the big idea. Here is how I count ID3 tag size :

  if (inbuf[0] == 'I' && inbuf[1] == 'D' && inbuf[2] == '3') //inbuf contains first 10 bytes from file
  
  {
  
      int size = inbuf[3] * 2097152 + inbuf[4] * 16384 + inbuf[5] * 128 + inbuf[6]; //Will change to binary shifts later
  
      //Do something else with it - skip rest of ID3 tags etc
  
  }

  It works ok for files without ID3 tags and for some files with them, but for some other files ffmpeg (which I use for decoding) returns "no header" error, which means it didn’t catch MP3 header correctly. I know that since if I remove ID3 from that .mp3 file (with Winamp for example), no errors occur. The conclusion is that size count algorithm isn’t always valid.

  So the question is : how do I get to know how big exactly is entire ID3 part of the .mp3 file (all possible tags, album picture and whatever) ? I’m looking for it everywhere but I just keep finding this algorithm I posted above. Sometimes also something about some 10 bytes footer I need to take into account, but it seems it frequently gets more than 10 bytes for it to eventually catch proper MP3 frame.