A Novel Framework for Multiple People Tracking Using Camera Networks with Overlapping Views

Abstract:

We present a novel framework for multiple pedestrian tracking using overlapping cameras in which the problems of object detection and data association are solved alternately. In each round of our algorithm, the people are detected by inference on a factor graph model at each time slice. The outputs of the inference, namely the probabilistic occupancy maps, are used to define a cost network model. Data association is achieved by solving a min-cost flow problem on the resulting network model. The outputs of the data association, namely the ground occupancy maps, are used to control the size of factors in graph model in the next round. By alternating between object detection and data association, a desirable compromise between complexity and accuracy is obtained. Our experiments involve challenging, notably distinct datasets and demonstrate that our method is competitive compared with other state-of-art approaches.

• Periodical: Applied Mechanics and Materials (Volumes 644-650)
• Pages: 1240-1244
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.644-650.1240
• Online since: September 2014
• Authors: A Chuan Li*, Jiu Qing Wan
• Keywords: Multi-Camera with Overlapping Views, Multi-People Tracking, Novel Framework

References

[1] S. Khan and M. Shah. IEEE Trans. on Pattern Analysis and Machine Intelligence. 2009, volI, no. 3, p.505-5l9.

[2] Jiang MingXin, Wang HongYu, Liu XiaoKai. Acta Automatica Sinica, 2012, 38(4): 531-539.

[3] Fleuret F, Berclaz J, Lengagne R, et al. Pattern Analysis and Machine Intelligence, 2008, 30(2): 267-282.

[4] Jiuqing W, Fan Z. Multi-camera people localization via cascaded optimization on higher-order MRFs. In: Proceedings of the Sixth International Conference on Distributed Smart Cameras. Hong Kong , China: IEEE, 2012. 1-7.

[5] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA: IEEE, 2005. 886-893.

DOI: 10.1109/cvpr.2005.177

[6] Lin H T, Lin C J, Weng R C. Machine learning, 2007, 68(3): 267-276.

[7] Kschischang F R, Frey B J, Loeliger H A. Information Theory, 2001, 47(2): 498-519.

[8] Berclaz J, Fleuret F, Turetken E, et al. Pattern Analysis and Machine Intelligence, 2011, 33(9): 1806-1819.

DOI: 10.1109/tpami.2011.21

[9] Information on: http: /cvlab. epfl. ch/data/pom.

[10] Information on: available: http: /pets2009. net.

[11] Kasturi R, Goldgof D, Soundararajan P, et al. Data, metrics, and protocol. Pattern Analysis and Machine Intelligence, 2009, 31(2): 319-336.