Hello list! I'm pondering a little thingy, a video sequence recognizer. I've thought about looking for variations in luminance and/or chrominance in the image, but come to the conclusion that my TV providers, at least, have none or very little consistency in their settings when broadcasting their stuff. One time the material may be murky and dark, other times slightly washed out, and other times still, oversaturated. I think it would be hard to recognize sequences under such conditions.
What, then, does not change between runs? The motion. No matter how they've messed up the video feed, short of mirroring it, the motion will still be the same. There is a layer in MPEG that deal with motion vectors. I could throw something together to obtain some sort of vectors, but it would most likely be slow, and unnecessary work at that. Would it be hard to separate the motion vector layer in, say, mpeg2enc, and get (possibly fingerprintable) chunks of motion vector sequences from it? /Sam ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Mjpeg-users mailing list Mjpeg-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mjpeg-users
