Transcoding vs. Scalability in Video Streaming for Heterogeneous Networks/Clients

Abstract

Transcoding provides ad hoc solutions for every user’s category by decompressing a video sequence and recompressing it into a different format where the frame size, colour representation, frame rate, quality, and other characteristics may change. Scalable coding allows implementing most of these operations in an easier and cheapest way, substituting decoding and re-encoding with simple selection and combination operations. In this case, the video is encoded once for all, but different users can access different instances of it according to their specific requirements, capabilities, and compatibilities. The aim of this chapter is to highlight the pros and cons of both techniques, presenting solutions based on layered coding and multiple description coding as a valuable alternative to transcoding, especially in those cases where network losses are not negligible.