Scalable AVS Video Codec System Staying In Motion

Video image is greatly favored in the field of communication because of the advantages such as intuition, tangibility, efficiency and universality, etc. But the massive of information greatly limits its processing and transmission in the Internet and mobile networks. Since1990s the internet and mobile communication technology is developing rapidly. Video compression attracts the attention of many scholars and made great achievement in video codec technology. In2006, the advanced audio and video coding standard AVS that owns independent intellectual property rights is implemented, which opens up a broad space for video communication technology development.As network technology becomes increasingly mature and more and more widely applied, storage oriented video codec has turned to network transmission oriented video codec at present. Storage oriented video codec is mainly restricted by bandwidth and terminal equipment diversification. How to make the network video service more flexible in order to meet the needs of different users and different terminal devices has become an urgent problem to be solved. This issue can be addressed with scalable video codec.The code stream acquired by compressing and encoding the scalable video encoding (SVC) for once can be decoded into video images of different bit rate, frame rate, spatial resolution and video quality with scalable decoder. Therefore, SVC is able to adapt to network and terminal devices, can provide a higher quality video within the prescribed rate range, and has the advantage of rate variable. However, good scalability can also cause higher computational complexity and increase the difficulty of practical application.The goal of this study is to design a scalable video codec system based on the above issues in order to reduce computational complexity of video codec as much as possible while improving video quality at the same time. Main work and innovation points are as follows:Based on a further study of the AVS codec principle and the codec reference software officially released, a scalable AVS video codec system staying in motion is proposed in this paper. Aiming at the disadvantage of high search complexity in traditional scalable coding, a scalable AVS video codec system staying in motion is designed in this study. This algorithm only carries out motion search in base layer, the corresponding enhancement layer use motion vector of base layer directly. Hence, decreasing motion search of enhancement layer not only reduces the bit of transmission but also decreases computational complexity greatly. The reduction of signal-to-noise PSNR caused by decreasing motion search will be compensated by a more precise filter and the relevant quantization table.