Motion compensated Prediction in Distributed Video Coding

The ongoing research in Distributed Video Coding (DVC) for low complexity encoding is trying to shorten the substantial performance gap to well known state-of-the-art coders. One of the reasons for this gap is the motion estimation without the reference frame. In this paper we look at what motion estimation to use. Furthermore we compare the two possible methods to get a motion compensated prediction. These are interpolation and extrapolation. Our results show that extrapolation outperforms the widely used interpolation noticeably.