Mining Video Associations for Efficient Database Management

To support more efficient video database management, this paper explores the concept of video association mining, with which the association patterns are characterized by sequentially associated video shots and their cluster information. Given a continuous video sequence V, the video shot segmentation mechanism is first introduced to parse it into discrete shots. We then cluster shots into visually distinct groups and construct a shot cluster sequence by using the class label of each shot. An association mining scheme is designed to mine sequentially associated clusters from the sequence. Those detected associations wi l l convey valuable knowledge for video database management. The experimental results demonstrate the effectiveness of our design.