AVSS '03 Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance
Tracking Multiple Humans in Complex Situations
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Novel Clustering-Based Method for Adaptive Background Segmentation
CRV '06 Proceedings of the The 3rd Canadian Conference on Computer and Robot Vision
ACM Computing Surveys (CSUR)
Tracking People in Crowds by a Part Matching Approach
AVSS '08 Proceedings of the 2008 IEEE Fifth International Conference on Advanced Video and Signal Based Surveillance
A survey on behavior analysis in video surveillance for homeland security applications
AIPR '08 Proceedings of the 2008 37th IEEE Applied Imagery Pattern Recognition Workshop
Fast Tracking of Humans in Frequently Occurring Entry, Exit and Occlusion Scenarios
ICCTD '09 Proceedings of the 2009 International Conference on Computer Technology and Development - Volume 02
Real time classification and tracking of multiple vehicles in highways
Pattern Recognition Letters
Robust Human Tracking Algorithm Applied for Occlusion Handling
FCST '10 Proceedings of the 2010 Fifth International Conference on Frontier of Computer Science and Technology
A survey on visual surveillance of object motion and behaviors
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
Object tracking with occlusion handling is a challenging problem in automated video surveillance. In particular, occlusion handling and tracking have been often considered as separate modules. This paper proposes a tracking method in the context of video surveillance, where occlusions are automatically detected and handled to solve ambiguities. Hence, the tracking process can continue to track the different moving objects correctly. The proposed approach is based on sub-blobbing, that is, blobs representing moving objects are segmented into sections whenever occlusions occur. These sub-blobs are then treated as blobs with the occluded ones ignored. By doing so, the tracking of objects has become more accurate and less sensitive to occlusions. We have also used a feature-based framework for identifying the tracked objects, where several flexible attributes were involved. Experiments on several videos have clearly demonstrated the success of the proposed method.