Human Tracking Based on Image Sequences

Human tracking is a challenging problem because of the non-rigid nature of the human body,and the frequent occlusion encountered among people.A new approach which is a combination of Kalman filter and Bayesian is presented to track multiple humans.First,it constructs a simple background model,extracts moving humans by using subtraction method,and constructs human model by using EM algorithm.Secondly,Kalman filter is used to track each human when there is no occlusion between humans;and Bayesian is used to segment and track each human when there is occlusion between humans.Experiments show that it can assure the rapidity of tracking,and also can deal well with the occlusion among people.The new approach is robust and the tracking results are satisfactory.