Real-time detection of sport in MPEG-2 sequences using high-level AV-descriptors and SVM

We present a new approach for classifying mpeg-2 video sequences as 'sport' or 'non-sport' by analyzing new high-level audiovisual features of consecutive frames in real-time. This is part of the well-known video-genre- classification problem, where popular TV-broadcast genres like cartoon, commercial, music video, news and sports are studied. Such applications have also been discussed in the context of MPEG-7 (1). In our method the extracted features are logically combined by a support vector machine (2) to produce a reliable detection. The results demonstrate a high identification rate of 98.5% based on a large balanced database of 100 representative video sequences gathered from free digital TV-broadcasting and world wide web. I. INTRODUCTION With the advent of digital TV-broadcasting, e.g. DVB and IPTV presenting more than hundreds of channels at a time, the need for a user-friendly TV-program selection is growing. Unlike the present situation, a new system should enable users to access programs clustered by genres according to Fig. 1.