An adaptive block-matching algorithm for motion estimation

Abstract

A new adaptive algorithm for the block matching motion estimation is presented. The algorithm works in the full-search fashion but unlike the FSBMA it adjusts the number of computations dynamically to picture variation. Due to incorporated mechanism of data-driven thresholding, the proposed approach performs as four times as less operations compared to the FSBMA while maintaining the same quality of results. Its hardware implementation is simple and compact. A supportive hardware design as well as simulation results on benchmarks are outlined.