H.264 Extension To The Standard - Scalable Video Coding (svc) And The Fine Granularity Scalable Coding Research

H.264 is the newest video coding standard established by two national video coding organizations: ITU-T and ISO/IEC. H.264 will make important sence in the areas as DVB, real-time video communication, network video streaming and multimedia message for its excellent coding efficiency and good network adaptability. ITU-T and ISO/IEC have established extending video coding standard—SVC (Scalable Video Coding) based on H.264. The goal of SVC is to provide scalability with good compression efficiency by allowing combinations of scalable layers, including the spatial layer, temporal layer and Signal-to-Noise Ratio (SNR) layer. This enables a simple and flexible solution for transmission over heterogeneous networks, additionally providing adaptability for bandwidth variations.This article introduced H.264 at the beginning. H.264 still uses prediction and transforming coding, and there is less change in coding frame as former standards. H.264 makes important changes with function models in details. The chapter two introduced the characteristic of H.264 and analyzed the key technologys of H.264 including DCT, intra-prediction, inter-prediction, entropy coding etc.Then it introduced the extending standard SVC. The core arithmetic of SVC is producing a base layer which has base quality and some enhancement layers which have additional informations through video compressing. Base layer can be decoded independently and enhancement layers depend on base layer. It could not be decoded accurately on the occasion that base layer data lost. The chapter three researched the performance of SVC by simulating based on the analyzing of its principle.At last it made a deep research of fine SVC. It makes more fine adaptability for bandwidth variations than traditional SVC methods. The chapter four introduced some arithmetics of fine SVC first, and then it analyzed principle of MGS (Medium Grain Scalability) the newst fine SVC method of the extending standard. Then it reseached performance of MGS through simulation, and proposed a new algorithm: AEL-MGS. The simulation shows that it improved the coding efficiency while achieved fine SVC.