Robust tracking algorithm using mean-shift and particle filter

Abstract

Aiming to the problems that Mean-Shift algorithm costs low computation, but easy to fall into local maximum, and huge computation of Particle Filter tracking algorithm leads to low real-time processing capacity, according to the need of real stereo vision measurement system, a kind of tracking algorithm which combines Mean-Shift and Particle Filter by essentiality function is proposed. Under the condition without occlusion, Mean-Shift is used to track object. When object is occluded, Particle Filter is applied to accomplish the later object tracking. These two algorithms alternate by a defined threshold. The tracking algorithm is used into real stereo vision measurement system. Experiment result indicates that the algorithm takes on high efficiency, so it is of high practicability.