PixSO: a system for video shot detection

In this paper, a system called PixSO (pixel-level comparison, segmentation and object tracking) is presented for effective shot change detection using an unsupervised object segmentation algorithm and the technique of object tracking based on the segmentation mask maps. The detection method was tested on TV news, commercial, sports and documentary video sequences which contain different types of shots having different object and camera motions. The results have shown that the PixSO system can not only produce accurate shot change detection, but also obtain object level information of the video frames, which is very useful for video content indexing and analysis in multimedia databases.

[1]  Rainer Lienhart,et al.  Comparison of automatic shot boundary detection algorithms , 1998, Electronic Imaging.

[2]  Wei Jyh Heng.  Shot boundary refinement for long transition in digital video sequence , 2002, IEEE Trans. Multim..

[3]  Rangasami L. Kashyap,et al.  Augmented Transition Network as a Semantic Model for Video Data , 2001 .

[4]  Atreyi Kankanhalli,et al.  Automatic partitioning of full-motion video , 1993, Multimedia Systems.

[5]  Ramesh C. Jain,et al.  Knowledge-guided parsing in video databases , 1993, Electronic Imaging.

[6]  Chong-Wah Ngo,et al.  On clustering and retrieval of video shots through temporal slices analysis , 2002, IEEE Trans. Multim..

[7]  Rangasami L. Kashyap,et al.  Indexing and searching structure for multimedia database systems , 1999, Electronic Imaging.

[8]  A. Murat Tekalp,et al.  Temporal video segmentation using unsupervised clustering and semantic object tracking , 1998, J. Electronic Imaging.

[9]  Dong-Seok Jeong,et al.  Detection of video scene breaks using directional informations in DCT domain , 1999, Proceedings 10th International Conference on Image Analysis and Processing.

[10]  Didier Le Gall,et al.  MPEG: a video compression standard for multimedia applications , 1991, CACM.

[11]  Young-Min Kim,et al.  Fast scene change detection using direct feature extraction from MPEG compressed videos , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[12]  Ramin Zabih,et al.  A feature-based algorithm for detecting and classifying scene breaks , 1995, MULTIMEDIA '95.

[13]  Boon-Lock Yeo,et al.  Rapid scene analysis on compressed video , 1995, IEEE Trans. Circuits Syst. Video Technol..