A New Resampling Strategy about Particle Filter Algorithm Applied in Monte Carlo Framework
Gang Wu, Zhenmin Tang · 2009
In this paper we propose a new resampling strategy about particle filter algorithm for tracking object in video sequence. We incorporate the new resampling strategy and adaptive elliptical template with the classical particle filter algorithm. We apply enhanced algorithm to track selected object in a standard video and demonstrate its performance compared with the algorithm proposed by K. Nummiaro. Experimental results show that the proposed particle filter algorithm improves the efficiency of tracking system, while it is unfluctuating even if the surroundings of visual tracking are under heavy fog.