Real-Time Tracking Based on Particle Filter and Multi-cue

This paper is a new attempt to introduce a ldquomulti-cue rdquo particle filter tracking method. It uses multiple features to facilitate 2D video tracking from a monocular view. The methodpsilas benefits lie in its robustness and speed. Speed is improved by adaptive adjustment of different cues; robustness is implemented by multiple cues. The proposed method is demonstrated on man tracking with a non-stationary camera in unconstrained outdoor environments. Result shows it is speedier and robust again background noise such as cluttered background, occlusion, color distracters and illumination change than conventional PF.