A Robust Visual Tracking Approach with Adaptive Particle Filtering

Abstract

In order to improve the robustness and stability as well as the computation efficiency of the video tracker based on particle filtering, an adaptive state evolution equation and an online increment learning observation likelihood model configured by an updatable eigen-basis of the object appearance subspace is combined into the particle filter to cope with the uncertainties during tracking, and the strategy of online self-adjusting the number of particle needed for approximating the state posterior density function is adopted to enhance the computation efficiency. The experimental results show that the approach proposed in this paper can not only track the moving object in the video reliably and effectively, but has nice robustness to the appearance variation caused by illumination, occlusion and pose changes.