On the Evolution of Videotext Description Scheme and Its Validation Experiments for MPEG-7

Videotext refers to text superimposed on still images and video frames, and can be used in many MPEG-7 applications that deal with archival and delivery of images and video. It can be used to annotate and index large video and image collections, and enables text based search, automatic video logging, and video cataloging. This paper describes the joint work of IBM and Philips Research Laboratories on designing an MPEG-7 description scheme based on videotext. It describes the elements comprising the Videotext DS and discusses validation experiments performed to demonstrate the effectiveness of the DS in video browsing and classification tasks.