Estimation and mode decision for spatially correlated motion sequences

There are several parts of the video encoding process that can be optimized for high quality, high compression ratio and fast encoding speed. In this paper we focus on some of these encoder optimization issues. We first study motion estimation, a computationally intensive part of encoding, and propose efficient algorithms based on exploiting spatially correlated motion, that perform well in terms of speed, quality and the bit rate for different kinds of motion sequences as well as picture formats. We extend our motion estimation algorithms to find motion vectors for sub-parts of a block, as allowed for in the H.263 and MPEG-4 standards. We study the one-four motion vector decision and propose an efficient two-tier decision strategy. We then study the error measurement or the measure of similarity that is an important parameter for motion estimation as well as the mode decision. We introduce a new error measurement that can improve the mode decision significantly. Furthermore for certain motion estimation strategies, this new measurement can also improve the bit rate resulting from a better motion estimation. Another issue that we address in this paper is the estimation of delta motion vectors in the H.263 and the MPEG-4 standards. We have implemented these optimizations in a H.263 framework and the results are encouraging.