HDH Based Compressed Video Cut Detection

This paper presents a video cut detection algorithm using multi-level Hausdor distance histograms (HDH). Hausdor distance is obtained by comparing edge points of successive frames, wherein the edge information is extracted from compressed frames directly. The use of Hausdor distance histogram instead of the comparison of entering/exiting edge pixel counts [9] makes the algorithm more robust to complicated camera shots. The extraction of edge information in compressed domain greatly escalates the cut detection process, which is critical for indexing of large amount of video materials in large scale video databases. The performance of this algorithm has been tested on a variety of videos. The experimental results show that this algorithm can robustly tolerate rapid changes in scene brightness as well as multiple object and camera motions. Bo Shen is with the Visual Computing Department, HP Laboratories, 1501 Page Mill Road, Palo Alto, CA 94304, U.S.A. Email: [email protected]