Color-Based Object Tracking Using a Hybrid Particle Filter
The efficient tracking of visual objects in complex environments is a challenging task for the vision community. In real-rime particle filter tracking, the particles are a constrained resource and should be well used. We propose a fast hybrid tracking method. At each time step, the procedure consists of a detection based particle filtering and a mean shift iteration followed. The detection based particle filter method efficiently explores the possible state space by evenly spreading particles and keeping a certain number of particles in each possible target region. The mean shift iteration is used to further improve the accuracy of state estimate. Color clues are used in this method.Experimental results show that when tracking rapidly moving targets, this method has better performance but need smaller particle number compared with standard particle filter method.
Tracking particle filter color clue mean shift
Yongfeng Cao Hui Yin Hong Sun
Signal processing laboratory School of Electronic Information Wuhan University Luoyu road, Wuhan 430079,China
国际会议
The International Colloquium on Onformation Fusion 2007(2007年国际信息融合研讨会)
西安
英文
324-330
2007-08-22（万方平台首次上网日期，不代表论文的发表时间）