Parallel Multi-Cue Tracking Based on Particle Filter

Abstract

This paper is a new attempt to introduce a "parallel multi-cue" particle filter tracking method. It combines multiple features distribution and multiple thread programming method to facilitate 2D video tracking from a monocular view. The method's benefits lie in its speed, its robustness. Speed is improved by parallel operation of multiple thread which dispose different cues at one time; robustness is implemented by multiple cues. The proposed method is demonstrated on head tracking with a non-stationary camera in unconstrained environments. Result shows it is speedier and robust again background noise such as cluttered background, occlusion, color distracters and illumination change than conventional single thread PF and single thread PF with multi-cue.