A 2-Stage Partial Distortion Search Algorithm for Block Motion Estimation

In this paper, we propose a novel 2-Stage Partial Distortion Search (2S-PDS) algorithm to reduce the computational complexity in block motion estimation algorithms. In this algorithm, an early-rejection stage is introduced where the partial distortion of a decimated pixel block is calculated and compared with its local minimum. A block is rejected without calculating the full distortion of the entire block if the partial distortion is larger. In order to reduce the probability of false rejection, the local minimum is amplified by a pre-defined threshold before the comparison. Experimental results show that the proposed algorithm can reduce the complexity of block motion estimation algorithm significantly with only marginal performance penalty. The proposed algorithm can be used in combination with full-search or other fast search algorithms.