Fast Motion Estimation using Hexagon-Based Search Pattern in Predictive Search Range

Abstract

The hexagon-based search pattern (HEXBS) algorithm yields fewer search points required for motion estimation, compared to square-shaped and diamond-shape patterns. In this paper, we propose a fast motion estimation algorithm to further reduce the search points demanded by HEXBS algorithm. After exploiting the statistical property of motion vectors of the neighboring blocks, the number of selected candidate points on the hexagon endpoints will be lower than the original HEXBS algorithm. Thus, the motion estimation efficiency can be improved in such a way. Experimental results show that the proposed algorithm decreases 57.62% average search points, compared to HEXBS algorithm, with only slight quality degradation.