Real-time shot change detection over online MPEG-2 video

The development of technologies and standards intended to enable and organize a variety of innovative services over the infrastructure that supports digital TV requires a parallel evolution of the techniques to manage its main information media: digital video. Particularly, as TV broadcasters confront the provision of dynamic value-added content-based services over the audio-visual information that is being transmitted, they require online annotation of the received digital channels. In order to achieve it, apart from high levels of success in video temporal segmentation-a great resource-consuming task previous to most video annotation techniques-a faster than real-time performance is crucial. This paper describes a software module for video temporal segmentation which is able to detect both abrupt transitions and all types of gradual transitions in real time, with levels of recall and precision not reported up to date for such a large and varied set of video sequences.