An adaptive block matching algorithm for motion estimation

In this paper a modified version of the predicted vector spectral search algorithm (PVSSA) is presented. The proposed method adaptively determines the search area for motion estimation. The maximum number of search points for finding the best matched block is lower than that of the PVSSA. Although the proposed algorithm is similar to the PVSSA in terms of regularity, it has lower computational complexity. To find the search area the temporal and spatial correlations among the motion vectors of the image blocks are exploited. Based on this, a matched block is chosen from a rectangular area that the prediction vectors set out. We have evaluated the algorithm through a comprehensive performance study that shows that the proposed algorithm achieves substantial speedup without quality loss for a wide range of video sequences, compared with other motion estimation techniques.