Ip camera using ffmpeg drawing on screen - c++

Question

I'm using ffmpeg 1.2 to take video from ip camera.I make it draw on the screen, so I wonder if there is some event mechanism to to know if it is time to call av_read_frame?
If I read frame not so frequent as the camera gives frames i get segmentation fault = on some malloc functions inside ffmpeg routines(video_get_buffer)
I also get segmentation fault just when drawing on screen.
In Render function call every 0 miliseconds
void BasicGLPane::DrawNextFrame()
{
int f=1;
while(av_read_frame(pFormatCtx, &packet)>=0)
{
// Is this a packet from the video stream?
if(packet.stream_index==videoStream)
{
// Decode video frame
avcodec_decode_video2(pCodecCtx, pFrame, &FrameFinished,
&packet);
// Did we get a video frame?
if(FrameFinished)
{
f++;
this->fram->Clear();
// if (pFrame->pict_type == AV_PICTURE_TYPE_I) wxMessageBox("I cadr");
if (pFrame->pict_type != AV_PICTURE_TYPE_I)
printMVMatrix(f, pFrame, pCodecCtx);
pFrameRGB->linesize[0]= pCodecCtx->width*3; // in case of rgb4 one plane
sws_scale(swsContext, pFrame->data, pFrame->linesize, 0, pCodecCtx->height, pFrameRGB->data, pFrameRGB->linesize);
//glGenTextures(1, &VideoTexture);
if ((*current_Vtex)==VideoTexture) current_Vtex = &VideoTexture2;else current_Vtex = &VideoTexture;
glBindTexture(GL_TEXTURE_2D, (*current_Vtex));
glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR );
glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR );
glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, pCodecCtx->width, pCodecCtx->height, GL_RGB, GL_UNSIGNED_BYTE, pFrameRGB->data[0]);
//glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB, pCodecCtx->width, pCodecCtx->height, 0, GL_RGB, GL_UNSIGNED_BYTE, pFrameRGB->data[0]);
//glDeleteTextures(1, &VideoTexture);
GLenum err;
while ((err = glGetError()) != GL_NO_ERROR)
{
cerr << "OpenGL error: " << err << endl;
}
// av_free(buffer);
}
}
// Free the packet that was allocated by av_read_frame
av_free_packet(&packet);
if (f>1) break;
}
//av_free(pFrameRGB);
}
The picture I get on the screen is strange (green quads and red lines are motion vectors of those quads)
http://i.stack.imgur.com/9HJ9t.png

Answer 1

If you hang on the video device, you should read frames as soon as camera has them; you should either query the camera hardware for supported FPS, or get this information from the codec. If no information is available, you have to guess.
It is suspicious that you get a crash when you don't read the frame in time; the worst which should happen in this case would be a lost frame.
Depending on the camera colorspace you may also need to convert it before showing it on a screen. I don't see you doing that.