A Novel Occlusion-Adaptive Object Tracking Method

Abstract

In conventional object tracking methods, much attention has been paid to tracking efficiency, but they often failed in tracking an occluded object. In this paper, we present a new method to improve the occlusion adaptability and tracking robustness. This proposed algorithm covers the occlusion-adaptive particle filter (OAPF) framework, which employs the adaptive state transition model to detect occlusions by a first-order histogram difference dynamic algorithm accurately and simply. Thus, when partial or complete occlusions occur, it can detect interrupted state transition to realize persistent tracking. In addition, tracking robustness is also upgraded via adaptive Gaussian noise coefficient model in particle propagation. Finally, we emphasize that the computing complexity of OAPF is evidently decreased by reducing the particle number in execution. As a result, this simple and effective occlusion-adaptive tracking method has been demonstrated through several real-time sequences.

Publication
2012 International Conference on Computer Science and Information Processing (CSIP)
Xiaofeng Lu
Xiaofeng Lu
PhD Student
Li Song
Li Song
Professor, IEEE Senior Member

Professor, Doctoral Supervisor, the Deputy Director of the Institute of Image Communication and Network Engineering of Shanghai Jiao Tong University, the Double-Appointed Professor of the Institute of Artificial Intelligence and the Collaborative Innovation Center of Future Media Network, the Deputy Secretary-General of the China Video User Experience Alliance and head of the standards group.