PERFORMANCE-BASED STUDY ON DIAMOND SEARCH AND ITS VARIATIONS FOR EFFICIENT MOTION ESTIMATION IN VIDEO

Abstract

The motion estimation is a crucial process of video encoding process which needs to be dealt with high importance. There has been a lot of research works on the topic and various block-based motion estimation techniques were designed and implemented for the same. Out of the existing developed techniques, Diamond Search (DS) is one of the efficient motion estimation technique and most no. of new algorithms are designed based on this DS only. In this paper, we will be critically reviewing the Diamond Search (DS) and few existing variations of block matching algorithms based on DS which varies according to search criteria, search pattern, size of the search window, macro block size and PSNR values. The algorithms are being applied to different video sequences to do a comparative study among them and check out which variation of DS is efficient for which video sequences.