A comparative study of block matching and optical flow motion estimation algorithms

Abstract

Block matching and optical flow algorithms are the two major motion estimation techniques that are widely employed today. The main aim of this paper is to compare the above two algorithms in terms of processing time, Peak Signal to Noise Ratio (PSNR), Structural Similarity (SSIM) and Mean Opinion Score (MOS). An exhaustive search block matching algorithm which has the maximum efficiency compared to any other block matching algorithm as well as the Brox's optical flow estimation algorithm are implemented. The algorithms are optimized by selecting appropriate parameter values that gives the best result. Then the algorithms are compared based on their motion estimated image for the same input image sequence and finally the results obtained are analyzed.