Columbia's Automatic Foreground Moving Video Object Segmentation Software

Contacts : Di Zhong and Shih-Fu Chang

This system has been built on top of our video segmentation software. Automatically segmented regions (i.e., low-level moving patches) are merged and tracked to detect foreground moving objects at the whole-body level (e.g., people, car), using motion fusion, spatio-temporal consistency filtering, and background boundary models.

Technical details can be found at the project home page.

Binary code is available in Java version or C++ version.

Publications:

D. Zhong and S.-F. Chang, Long-Term Moving Object Segmentation and Tracking Using Spatio-Temporal Consistency, IEEE International Conference on Image Processing, Greece, Oct. 2001.
(PS.GZ/PDF)

D. Zhong and S.-F. Chang, Video Object Model and Segmentation for Content-Based Video Indexing, IEEE International Symposium on Circuits and Systems (ISCAS'97), Hong Kong, June 1997, Special Session on Networked Multimedia Technology and Application.
(PS.GZ/PDF)

D. Zhong and S.-F. Chang, An Integrated Approach for Content-Based Video Object Segmentation and Retrieval, IEEE Transactions on Circuits and Systems for Video Technology, Vol. 9, No. 8, pp. 1259-1268, Dec. 1999.
(PS.GZ/PDF)

For problems or questions regarding this web site contact The Web Master.
Last updated: June 12, 2003.