A fully SNR, spatial and temporal scalable 3DSPIHT-based video coding algorithm for video streaming over heterogeneous networks

Video streaming over heterogeneous networks (e.g. the Internet) requires high degree of scalability from the video coding. To achieve all types of scalability required for a fully scalable video coding, a modification of the 3DSPIHT algorithm is presented in this paper. The proposed algorithm, called fully scalable 3DSPIHT (FS-3DSPIHT), adds spatial and temporal scalability features to the 3DSPIHT algorithm, through the introduction of resolution dependent lists and a resolution-dependent sorting pass. The important features of the original 3DSPIHT coder such as compression efficiency and full embeddedness are kept. The output bitstream of the FS-3DSPIHT encoder consists of a set of embedded parts related to the various qualities, spatial and temporal resolutions. It can be easily reordered and truncated by a simple parser in order to adapt various clients' requirements (e.g. quality, spatial resolution and temporal resolution) as well as bandwidth variation in a heterogeneous network.