Everything You Wanted to Know About MPEG-7: Part 1

65 Part I of this article provides an overview of the development, functionality, and applicability of MPEG-7. We'll first present the role of MPEG-7 within the context of past MPEG standards. We then outline ideas of what should be possible using MPEG-7 technology. In Part II, we'll discuss the description of MPEG-7's concepts, terminology, and requirements. We'll then compare MPEG-7 to other approaches on multimedia content description. B ecause of the Internet's popularity, the last decade has experienced a rapid increase of digital audiovisual information. Though the increasing availability of potentially interesting information has enriched our lives (for example, e-mail and the World Wide Web), the overwhelming amount of information also raises a fundamental problem: How fast and easily can desirable information be made available? The more interesting—that is, specific and useful—material available, the harder it is to locate. A noticeable indicator of the existing tension between humans and the vast amounts of information available lies in the popularity of search engines available on the Web. Unfortunately, current solutions let users only search for textual information. Identifying audiovisual information proves difficult, as no generally recognized description of this material exists. In general, it's not possible to efficiently search the Web for, say, a picture of the motorbike from " Terminator II, " or for a sequence where King Lear congratulates his assistants on the night after the battle, or for " twenty minutes of video according to my preferences of today. " You can envisage similar examples for audio, in which you whistle a melody to find a song or quote a movie to find the context. It's true that in specific cases, solutions do exist. Multimedia databases on the market today let users search for pictures using characteristics like color, texture, and information about the shape of objects in the picture. Furthermore, the question of identifying content isn't restricted to database retrieval applica-tions—the problem applies equally to other areas. For instance, we're promised a world of 500-plus broadcast television channels, which will of course make it harder to select a potentially interesting channel. Domains other than search or filtering include image understanding (surveillance, intelligent vision, smart cameras, and so on) or media conversion (speech to text, picture to speech, visual transcoding, and so on). In October 1996, the Moving Pictures Expert Group (MPEG) started a new work item to provide a solution to the questions described above. The newest …