H.264 Algorithm Optimization

The newest video coding standard H.264, jointly made by ITU-T Video Coding Experts Group (VCEG) and ISO/IEC MPEG video group, has gained more and more attention recently, mainly due to its high coding efficiency and minor increase in decoder complexity.It can provide both objective and subjective image quality superior to existing standards. The main goals of this standardization are to develop a simple and straightforward video coding design with enhanced compression performance. Compared to MPEG-4 advanced simple profile, up to 50% of bit-rate reduction can be achieved.Some new techniques, such as spatial prediction in intra coding, adaptive block size motion compensation with small block sizes, Quarter-pel accurate motion compensation, 4x4 integer transformation, synchronization/switching pictures, multiples reference pictures and context adaptive binary arithmetic coding (CABAC), are used in this standard.With the use of these techniques, both compression efficiency and encoding quality are greatly improved. The testing results show that H.264 can greatly outperform existing video coding standards in PSNR, visual quality and compression efficiency.However, most of the coding tools adopted in H.264 involve very large computation complexity. Therefore the complexity of the encoder and computational cost are increased at the same time. That makes it very difficult for practical applications such as in real time video communication application or in some embedded system.To reduce the complexity of H.264, a number of efforts have been made to explore the fast algorithms in motion estimation, intra mode prediction and inter mode prediction for H.264 video coding.To overcome this defect, many fast BMAs are developed such as Diamond Search (DS), Hexagon Based Search (HEXBS), Three Step Search (TSS), Hierarchical Block matching Algorithms (HBMA), Cross Search (CS), Block-based Gradient Descent Search (BBGDS) and Unsymmetrical-Cross Muti-Hexagon Search(UMHexagonS).This paper presents two fast algorithms, 4-T fast algorithm and a table-search...