Multiple object tracking with energetic particle filtering and GVF-snake

A novel adaptive multi-object tracking algorithm with modified particle filtering and GVF-snake is proposed. Through the combination of modified particle filtering and GVF-snake, a novel energetic particle filtering (EPF) object tracking algorithm is proposed; and the multiple objects can be tracked with modified K-means clustering and energetic particle filtering (EPF). The tracking tactic for partially occluded objects is also proposed to overcome the effects such as occlusions in a complex environment. Experiments show that the proposed algorithm can fully utilize the information of mostly non-occluded object contour points in the entire tracking process, and obtain better tracking results even though the tracked objects are occluded for long time.