The Avs And H.264 Interpolation Algorithm And Weighted Prediction Part Of The Soc Implementation

With the people's demands for video becoming more criticl and the application of multimedia video compression technique growing rapidly, Video compression technique has becomes the emphasis of the high-tech field. H.264/AVC is the newest video coding standard of the Joint Video Team(JVT), which is organized by the ITU-T Video Coding Experts Group(VCEG) and the ISO/IEC Moving Picture Experts Group(MPEG).Compared with the former standard, H.264/AVC has been made great improvement in computing accuracy and algorithm. AVS is the video coding standard of research center which organized by Chinese Academy of Sciences. Compared with the H.264/AVC it has same display effect and less decoding complexity and area. Both of the two standards have broad foreground. Because the decoding complexity of H.264/AVC and AVS is very high, this paper adopts the SOC (System-on-a-Chip) to realize the decoder for AVS and H.264/AVC.This paper analyzes the detail of AVS and H.264/AVC, and then according to each standard feature it makes HW/SW partition and explanation. This paper uses Top-down design method to design the interpolation and weighted prediction in AVS and H.264/AVC. In order to satisfy real-time decoding, it adopts pipeline parallel structure to improve compute speed. Base on it, this paper analyzes reconfigurable structure of the luminance and chroma sub-pixel, and adopts reconfigurable structure for the chroma sub-pixel.This paper designs C model to verify decoding algorithms and provide test vector. And completes hardware design, verilog description, function and FPGA verification. The design is synthesized based on 0.18μm SMIC18 library. The highest frequency is 200 MHz. The design can decode the High-Definition video in real-time for AVS and H.264/AVC.