Scalable Video Compression
Report a problem or upload filesIf you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.
Scalable media compression algorithms are desirable because they allow the content to be compressed without prior knowledge of the set of bit‐rates and/or resolutions at which it is to be distributed and decoded. Scalable image compression technologies are now well understood and highly competitive with non‐scalable variants. In fact, applications that rely upon the scalability and interactive accessibility features of the highly scalable image compression standard JPEG2000 have been expanding in recent times and some of these applications intersect with the domain of video compression. Highly scalable video compression itself is fundamentally more challenging than scalable image compression, primarily because efficient video compressors rely upon the explicit estimation and communication of motion side information. Nevertheless, important advances in scalable video compression have been made over the past decade. One outcome of such developments is the SVC extension of H.264/AVC, where multiple spatial and temporal resolution layers are compressed with their own motion fields, but with the aid of inter‐layer prediction. Other approaches, replace prediction alone with motion compensated 3D wavelet transforms, with approximately orthogonal basis functions. From a theoretical perspective, such approaches are fundamentally superior to interlayer prediction; moreover, they have the property that all information associated with lower resolutions is fully embedded inside the higher resolution information, which can have important advantages in interactive browsing applications. However, 3D wavelet transforms have their own drawbacks, primarily related to the representation and scaling of suitable motion fields. This presentation will present an overview of the major approaches that have been taken to the problem of scalable video compression, including some of the more important theoretical concepts. Various applications for scalable video coding will be presented, including emerging applications. The talk will highlight what has been achieved, as well as future directions for research and some alternate paradigms for exploiting redundancy that may compete with scalable video compression. Considering the importance of motion, the presentation will also provide some insights and recent results related to new approaches to the efficient and scalable representation of motion information.
Link this pageWould you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !