A fast block motion estimation using progressive partial distortion search

Real-time video applications, such as the speed-oriented videoconferencing require fast motion estimation while high quality video entertainments require motion estimation with small prediction error. In this paper, a novel block-matching algorithm for fast motion estimation named progressive partial distortion search algorithm (PPDS) is proposed. It uses a halfway-stop technique with progressive partial distortion (PPD) to increase the chance of early rejection of impossible candidate motion vectors at very early stages. Experimental results on different kinds of PPD show that PPDS provides a wide range of computational reduction from 19.37 to 62.39 times, with less than 1 dB degradation on PSNR performance as compared to the full search algorithm. It shows PPDS is suitable for a wide range of video applications. The proposed PPDS gives a compromised computational reduction from 28 to 38 times with only 0.34-0.50 dB degradation on PSNR performance as compared to the full search algorithm.