Realtime Robust Object Tracking

Visual tracking is an important and active research area in computer vision. Usually a similarity measure between two probability distributions is used for tracking. The commonly used similarity measures such as Kullback-Leibler distance and Bhattacharyya distance are limited to one or two feature dimensions, due to the difficulty in estimating the entropy of the high-dimensional features. We proposed a similarity measure which is the sum of all pair-wise kernelized distances between two distributions. We used our similarity measure and mean-shift technique to track single object in a joint feature-spatial space, and achieved more accurate and robust tracking performance.

For multiple object tracking, we used distinctive features and particle filter framework to simultaneously and reliably track them. The quasi-random sampling and efficient distinctive features are used to achieve a realtime tracking speed.

References

1. C. Yang, R. Duraiswami and L. Davis. Efficient Spatial-Feature Tracking via the Mean-Shift and a New Similarity Measure. To appear in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Diego, CA, June 2005.
2. C. Yang, R. Duraiswami and L. Davis. Robust and Efficient Object Tracking Based on the Particle Filter. Submitted for publication, 2005.
3. B. Han, C. Yang, R. Duraiswami and L. Davis. Bayesian Filtering and Integral Image for Visual Tracking. Invited to special session of Real-Time Object Tracking: Algorithms and Evaluation in Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS), Montreux, Switzerland, 2005.
4. C. Yang, R. Duraiswami, A. Elgammal and L. Davis. On-Line Kernel-Based Tracking in Joint Feature-Spatial Spaces. DEMO on IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004.
5. C. Yang, R. Duraiswami, A. Elgammal and L. Davis. Real-Time Kernel-Based Tracking in Joint Feature-Spatial Spaces. Technical Report CS-TR-4567, Dept. of Computer Science, University of Maryland, College Park, 2004.

Some sequences in the paper [1]:
1. Ball sequence. (1.4MB)
2. Walking sequence. (1.1MB)
3. Predator sequence. (8.1MB)
4. Pedestrain sequence. (1.1MB)