Fractional-pel motion refinement based on hierarchical adjustable dual-parabola model

H.264 is the new video standardization recommended by the ITU-T Joint Video Coding Experts Group (JVT), in which fractional pixel motion compensation is applied in order to achieve more accurate motion description and higher compression efficiency. It, however, remarkably increased complexity in motion estimation process. Therefore reducing the computation load incurred by fractional-pel search is necessary. In this paper, an efficient fast fractional-pel motion refinement algorithm is proposed based on a hierarchical adjustable dual-parabola model (HADP). The significant fractional-pel positions can be accurately selected based on the adaptive HADP model, which greatly eliminates the unnecessary fractional-pet positions.