A MULTIATTRIBUTE SHOT SEGMENTATION ALGORITHM FOR VIDEO PROGRAMS

This paper describes a new algorithm for detecting cuts, thereby segme nting a video into shots. Our Web-based video library contains a large volume of news and documentary material; most of the transitions betw een shots in that type of programming are cuts, rather than dissolves or other complex transitions. We have developed an accurate multi-attr ibute algorithm for detecting cuts in video programs. The algorithm us es a motion metric to identify a set of cuts, then uses luminance hist ograms to eliminate false cuts. Our experimental results show that thi s algorithm is more accurate than previous motion-based transition det ection algorithms.