Mean-Shift Guided Particle Filter Tracking Method Based on Layered Dynamic Template Update

Abstract:

Traditional tracking algorithm is not compatible between robustness and efficiency, under complex scenes, the stable template update strategy is not robust to target appearance changes. Therefore, the paper presents a dynamic template-update method that combined with a mean-shift guided particle filter tracking method. By incorporating the original information into the updated template, or according to the variety of each component in template to adjust the updating weights adaptively, the presented algorithm has the natural ability of anti-drift. Besides, the proposed method cope the one-step iteration of mean-shift algorithm with the particle filter, thus boost the performance of efficiency. Experimental results show the feasibility of the proposed algorithm in this paper.

• Periodical: Advanced Materials Research (Volumes 846-847)
• Pages: 1217-1220
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.846-847.1217
• Online since: November 2013
• Authors: Yuan Zheng Li*
• Keywords: Mean-Shift, Object Tracking, Particle Filter (PF), Template Update

References

[1] Katja N, Esther K M, Luc V G. An adaptive color-based filter, [J]. Image Vision Computing, 2003, 21(1). pp.99-110.

[2] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using mean shift, [C]. Computer Vision and Pattern Recognition. Hilton Head, SC, USA: IEEE Computer Society, pp.142-149, June (2000).

DOI: 10.1109/cvpr.2000.854761

[3] Shan Cai-feng, Wei Yu-cheng, Tan Tie-niu, Ojardias F. Real time hand tracking by combining particle filtering and mean-shift, [A]. Sixth IEEE International Conference on Automatic Face and Gesture Recogniton[C]. Seoul, Korea: IEEE Computer Society, 2004. pp.669-674.

DOI: 10.1109/afgr.2004.1301611

[4] Babu R V, Perez P, Bouthemy P. Robust tracking with motion estimation and local kernel-based color modeling,. Image and Vision Computing, 2007, 25(8): pp.1205-1216.

DOI: 10.1016/j.imavis.2006.07.016

[5] Schreiber D. Robust template tracking with drift correction,. Pattern Recognition Letters, 2007, 28(12): pp.1483-1491.

DOI: 10.1016/j.patrec.2007.03.007

[6] Porikli F, Tuzel O, Meer P. Covariance tracking using model update based on Lie algebra., In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2006. pp.728-735.

DOI: 10.1109/cvpr.2006.94

[7] Matthews I, Ishikawa T, Baker S. The template update problem,. IEEE Trans on Pattern Analysis and MachineIntelligence, 2004, 26(6): pp.810-815.

DOI: 10.1109/tpami.2004.16

[8] Information on www. research. ibm. com/peoplevision.