Efficient partial distortion search algorithm for block based motion estimation

Abstract

The block based full search algorithm has been widely used for motion estimation in video coding, but it has the serious problem of significant computation requirements. In order to reduce the computation, this paper proposes a novel partial distortion search algorithm which reduces the computation of each distortion measure by using partial distortion search. In this algorithm, the entire macroblock is divided into different sub-blocks and the calculation order of partial distortion is determined based on the complexity of sub-blocks. A lossy algorithm is also presented by adaptively changing the early termination threshold for the current accumulated partial sum of absolute difference value. Experimental results show that the proposed lossless and lossy algorithm can significantly save about 60% and 70% of the total computational costs, respectively. PSNR degradation of lossy algorithm is very negligible and 0.016 dB on average.