Interactive adaptive movie annotation

the visual content of movies is essential for annotation. We present the interactive and adaptive i-Notation system, which describes actors’ names, automatically processes multimodal information sources, and deals with available sources’ varying quality. It provides the basis for intelligent interaction and demonstrates significant improvements in annotation efficiency. W ith the advance of digital video, movie viewers gain more control over what they see. We expect much more interactivity in films produced for DVD systems and online entertainment-on-demand systems.1 A likely application is nonlinear video browsing (see Figure 1), letting viewers jump to their favorite scenes, actors, jokes, and so on. Rather than ask for a predefined subject, viewers may want to describe their interests. Hence, future interactive movie systems need to deal with a wide variety of requests. Our goal is to assist creators of interactive movie applications by enriching the video data with semantic metadata.