Chapter 10: Content-Based Video Coding (MPEG-4)

Overview

MPEG-4 is another ISO/IEC standard, developed by MPEG (Moving Picture Experts Group), the committee that also developed the Emmy Award winning standards of MPEG-1 and MPEG-2. While MPEG-1 and MPEG-2 video aimed at devising coding tools for CD-ROM and digital television respectively, MPEG-4 video aims at providing tools and algorithms for efficient storage, transmission and manipulation of video data in multimedia environments [1,2]. The main motivations behind such a task are the proven success of digital video in three fields of digital television, interactive graphics applications (synthetic image content) and the interactive multimedia (worldwide web, distribution and access to image content). The MPEG-4 group believe these can be achieved by emphasising the functionalities of the proposed codec, which include efficient compression, object scalability, spatial and temporal scalability, error resilience etc.

The approach taken by the experts group in coding of video for multimedia applications relies on a content-based visual data representation of scenes. In content-based coding, in contrast to conventional video coding techniques, a scene is viewed as a composition of video objects (VO) with intrinsic properties such as shape, motion and texture. It is believed that such a content-based representation is a key to facilitating interactivity with objects for a variety of multimedia applications. In such applications, the user can access arbitrarily shaped objects in the scene and manipulate these objects.

The MPEG-4 group has defined the specifications of their intended video codec in the form of verification models (VM) [3]. The verification model in MPEG-4 has the same role as the reference and test models defined for H.261 and MPEG-2, respectively. The verification model has evolved over time by means of core experiments in various laboratories round the world. It is regarded as a common platform with a precise definition of the encoding and decoding algorithms that can be represented as tools addressing specific functionalities of MPEG-4. New algorithms/tools are added to the VM and old algorithms/tools are replaced in the VM by successful core experiments.

So far the verification model has been gradually evolved from version 1.0 to version 11.0, and during each evolution new functionalities have been added. In this Chapter we do not intend to review all of them, but instead to address those functionalities that have made MPEG-4 video coding radically different from its predecessors. Hence, it is intended to look at the fundamentals of new coding algorithms that have been introduced in MPEG-4. Before going into details of this coding technique, let us examine the functionalities of this codec.