MPEG-4: a multimedia standard for the third millennium. 2

Abstract

M PEG-4 defines a multimedia system for interoperable communication of complex scenes containing audio, video, synthetic audio, and graphics material. In part 1 of this two-part article (in the previous issue) we provided a comprehensive overview of the technical elements. Here in part 2 we describe an application scenario based on digital satellite television broadcasting, discuss the standard's envisaged evolution, and compare it to other activities in forums addressing multimedia specifications. The first MPEG-4 demo over satellite In April 1997 the Systems subgroup founded an implementation ad hoc group (IM1) with the mandate of validating the concepts of the MPEG-4 standard and promoting it by developing demonstration applications. This group also contributes to part 5 of the standard, the reference software. A system architecture was developed and successfully demonstrated at the 45th MPEG-4 meeting, where the world's first real-time display of MPEG-4 content (multicast over satellite) was shown to the MPEG community and the world press. Here we give a concise description of the architecture and demo mock-up. MPEG-4 broadcast scenario The scenario is an electronic programming guide (EPG) application where the user can choose among channels providing CD-quality audio, films, and news. The content is premultiplexed and then streamed. The target band for the overall application is 5 megabits per second. The overall system is constructed of a single (logical) origination point, a real-time, unidirectional communication channel, and a large number of end-user receiver/decoder terminals. It is a oneto-many, or possibly a few-to-many, system. The asymmetry of the architecture leads to an emphasis on reducing complexity and cost at the receiving side even if this implies increasing the complexity and cost at the transmitting side. The architecture is schematically depicted in Figure 1. The demonstration consisted of MPEG-4 server components located in Sunnyvale, California linked via a hybrid satellite network to MPEG-4 clients in Atlantic City, New Jersey. MPEG-4 takes advantage of the latest compression technologies, providing rendering of multiple objects within one viewing window. The network uses the Internet Group Multicast Protocol (IGMP) to route 76

Reducing 3D Fast Wavelet Transform Execution Time Using Blocking and the Streaming SIMD Extensions

“…Furthermore, the presented techniques could be generalized to other multimedia applications based on the computation of a transform such as JPEG-2000 [5], MPEG-2 [39] (based on the DCT transform) or MPEG-4 [12,13] on general-purpose processors. All transforms follow a common computation model.…”

“…Since one of the three spatial dimensions can be considered similar to time, a 3D subband coding using the zerotree method (EZW) was presented to code video sequences [10] and posteriorly improved with an embedded wavelet video coder using 3D set partitioning in hierarchical trees (SPIHT) [11]. Today, the standard MPEG-4 [12,13] supports an ad-hoc tool for encoding textures and still images based on a wavelet algorithm.…”

Semantic multimedia remote display for mobile thin clients

“…Under the framework of MPEG-4 standard, a dedicated multimedia scene description language, called Binary Format for Scene (BiFS) [19], [20] is defined. It describes the heterogeneous content of the scene, manages the scene object behavior (e.g.…”

Intelligent Semantic Access to Audiovisual Content

“…The focus of technological attempts has been on the analysis of digital video, due to its large amounts of spatio-temporal interrelations, which turns it into the most demanding and complex data structure. Current and evolving international standardization activities, such as of the EBU, MPEG-4 [3,4,9], MPEG-7 [5][6][7][8], or JPEG-2000 [10] for still images, deal with aspects related to data structures and metadata. In particular, the new MPEG standards are object-oriented, i.e., adopt video objects as the information units, which is different from the information units used in the current form of video and film, i.e.…”