9. Switched Scalable Video Coding Systems

Abstract

Scalable video coding means that a video sequence is only encoded once and the coded bitstream has the capability of sending any sub-bitstream relevant to a demand/request. In other words, the required sub-bitstream is simply read and packetized from the compressed bitstream. A video coding system that needs to know the scalable-point of the demands prior to the coding process or one that achieves scalability by transcoding is not truly scalable. Certainly, the coding efficiency can be improved by using the priori knowledge of the desired scalable-point in a scalable video coding system.