Study On Fast Mode Decision Algorithm For H.264/AVC Video Coding Standard

H.264 is the newest international video coding standard. It was developed by the JVT(Joint video team)which is composed of MPEG(moving picture expert group) of ISO/IEC and VCEG(video coding expert group)of ITU-T. H.264 focuses on achieving high coding effciency and improving reconstruct image quality. But it has a very high computational complexity, and it has limited its application. Therefore, reducing the computational complexity becomes its research focus.In this paper, first of all, we have a deep study for H.264, summarize the coding theory and describe its new techniques. Then, we have a deep analysis and research for the selection method of the best prediction mode. And on this basis, the fast and efficient prediction mode selection algorithms which have low complexity are proposed in this paper.A fast and efficient intra-frame prediction mode selection algorithm is proposed in this paper. Firstly, it makes a chose between the INTRA₄×4 and INTRA₁6×16 based on the complexity of the image. And then it adopts the high correlation of the prediction mode between current 4×4 block and adjacent blocks, strong correlation between suns of absolute transform difference (SATD) and rate distortion (RD) performance to skip some unlikely prediction modes efficiently. So it reduces the intra prediction complexity greatly. Experimental results show that the algorithm proposed can save the encoding time up to 65.5% averagely without noticeable loss of coding efficiency.In inter-frame prediction part, we offer two methods. The first one, it adopts the membership function to determine the characteristics of the macroblock, thus narrows the range of the alternative modes. The second one, its main idea is to use the technique skipping unnecessary modes. The percentage of 16×16 block size modes is the largest in most of the sequences. This means that many redundant mode calculations can be removed. Experimental results show that these two algorithms proposed all can achieve up to 20% speed up ratio with a little PSNR loss. Increase of total bits encoded is also not much noticeable.