A New Tracking Technique: Object Tracking and Identification from Motion

Pattern recognition and object tracking play very important roles in various applications, such as motion capture, object detection/recognition, video surveillance, and human computer interface. One very useful method that is rarely mentioned in literature is performing recognition from the motion cue. In many situations, the motion of an object is very representative and informative; therefore, it is possible to identify the object and its behavior from its motion. In this paper, we propose an original method to both identify and track an object in dynamic scenes. The method works on the situations with occlusions, appearance changes and global camera motions. It does not require prior segmentation or initialization. We test this method on a video database containing 18 World Cup soccer videos recorded from TV to detect and track the soccer ball. The results are satisfying. The results are also integrated into a video indexing system and the improvement on video retrieval is described.

[1]  Shih-Fu Chang,et al.  Algorithms and system for segmentation and structure analysis in soccer video , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[2]  G C Dean,et al.  An Introduction to Kalman Filters , 1986 .

[3]  Mei Han,et al.  Baseball scene classification using multimedia features , 2002, Proceedings. IEEE International Conference on Multimedia and Expo.

[4]  Adam L. Berger,et al.  A Maximum Entropy Approach to Natural Language Processing , 1996, CL.

[5]  Ying Wu,et al.  A co-inference approach to robust visual tracking , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[6]  Shih-Fu Chang,et al.  Structure analysis of sports video using domain models , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[7]  Jitendra Malik,et al.  Motion segmentation and tracking using normalized cuts , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[8]  Chong-Wah Ngo,et al.  On clustering and retrieval of video shots , 2001, MULTIMEDIA '01.

[9]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.

[10]  Ichiro Ide,et al.  An object detection method for describing soccer games from video , 2002, Proceedings. IEEE International Conference on Multimedia and Expo.

[11]  HongJiang Zhang,et al.  Automatic parsing of TV soccer programs , 1995, Proceedings of the International Conference on Multimedia Computing and Systems.

[12]  Shih-Fu Chang,et al.  Automatic selection of visual features and classifiers , 1999, Electronic Imaging.

[13]  Gudula Retz-Schmidt,et al.  A REPLAI of SOCCER: Recognizing Intentions in the Domain of Soccer Games , 1988, ECAI.

[14]  Jean Ponce,et al.  Computer Vision: A Modern Approach , 2002 .

[15]  Richard J. Qian,et al.  Detecting semantic events in soccer games: towards a complete solution , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[16]  Shmuel Peleg,et al.  Panoramic mosaics by manifold projection , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[17]  Yves Jean,et al.  Ball tracking and virtual replays for innovative tennis broadcasts , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.