Segmentation of players and team discrimination in soccer videos

In this paper, we introduce methods to extract low-level features for soccer video analysis. A new method is proposed to segment players by using a mean distributed color feature. In order to discriminate which team the player belongs to, we use mutual chromatic correlation degree of players to identify team without extracting templates of players in advance. Experimental results are included to show the effectiveness of the method.

[1]  Shih-Fu Chang,et al.  Real-time view recognition and event detection for sports video , 2004, J. Vis. Commun. Image Represent..

[2]  A. Murat Tekalp,et al.  Automatic Soccer Video Analysis and Summarization , 2003, IS&T/SPIE Electronic Imaging.

[3]  Jun-ichi Hasegawa,et al.  Development of motion analysis system for quantitative evaluation of teamwork in soccer games , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[4]  Alberto Del Bimbo,et al.  Semantic annotation of soccer videos: automatic highlights identification , 2003, Comput. Vis. Image Underst..

[5]  Alberto Del Bimbo,et al.  Taking into Consideration Sports Semantic Annotation of Sports Videos Content-based Multimedia Indexing and Retrieval , 2002 .

[6]  Hanspeter Bieri,et al.  SoccerMan-reconstructing soccer games from video sequences , 1998, Proceedings 1998 International Conference on Image Processing. ICIP98 (Cat. No.98CB36269).

[7]  Shih-Fu Chang,et al.  Structure analysis of soccer video with domain knowledge and hidden Markov models , 2004, Pattern Recognit. Lett..

[8]  Kazuyoshi Yoshino,et al.  Qualitative image analysis of group behaviour , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.