We are a small company producing a video analytics product which will ingest MPEG4 & MJPEG video from an IP camera and feed the decoded frames into our analytic algorithms. We've been looking over the IPP library and think it highly suitable to our project.
We've been looking over the audio-video samples and especially the UMC, but don't see anything that directly addresses the problem of ingesting a network stream, decoding it and then feeding individual frames to something other than a renderer. Does anyone have any examples of such a process that they would be willing to share? I would think that this would be a fairly common application of the image processing part of the IPP library. We've got the network access parts pretty much handled, it's what comes downstream of that that's giving us headaches.
N.B. The application will be running under Linux.