Abstract: Most object tracking are based on video sequence, visual object tracking is often disturbed by illumination change and the clutter background, it makes the result is of poor robustness. In order to obtain stable tracking results, this paper uses depth data for tracking. Taking into account the stability of the depth data, it does not vary with the illumination or color change, in this paper, we design an object tracking algorithm based on color and depth data fusion. To track object, an integrated particle filtering method is implemented. Due to use depth data in tracking, it shows the trajectory of the object at the same time. Experimental results show that this fused formation algorithm can achieve strong robustness, and have stability result from illumination change and similar color, and also exactly track target in complex scene.

Key words: object tracking, depth camera, particle filtering