Fragment-based variational visual tracking

Yi Zhou (1) , Hichem Snoussi (1) , Shibao Zheng (2) , Cédric Richard (1) , Jing Teng (1)

1 LM2S - Laboratoire Modélisation et Sûreté des Systèmes
2 Shanghai Jiao Tong University [Shanghai]

We propose a Bayesian tracking algorithm based on adaptive fragmentation and variational approximation. By using the cue of gradient, we fragment the target into disconnected rectangles and reduce the confusion from the background. To handle the uncertainties in real tracking case, we choose the Bayesian framework with a variational implementation. The parameters of the variational inference are updated according to the observation and to the weights of the voting candidates. Experimental results show that our tracker outperforms directive searching and particle filtering. Furthermore, due to the simplicity of calculation, the proposed method can be applied to real-time surveillance systems.

Domains: Engineering Sciences [physics]; Signal and Image processing

Yi Zhou, Hichem Snoussi, Shibao Zheng, Cédric Richard, Jing Teng. Fragment-based variational visual tracking. 2009 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Dec 2009, Aruba, Netherlands. pp.376-379, ⟨10.1109/CAMSAP.2009.5413252⟩. ⟨hal-02317801⟩