Multiple Persons Tracking with Data Fusion of Multiple Cameras and Floor Sensors Using Particle Filters

Abstract. Successful multi-target tracking requires locating the targets and labeling their identities. For the multi-target tracking systems, the latter becomes more challenging when the targets frequently interact with each other. In this paper, we propose a method for multiple per-sons tracking using multiple cameras and floor sensors. Our method esti-mates 3D positions of human body and head, and labels their identities. The method is composed of multiple particle filters that interact only in the exclusion occlusion model. Each particle filter tracks each person correctly by integrating information from floor sensors and the target-specific information from multiple cameras. Integration of these two types of sensors enables complement of each weak point and the correct track-ing of the target. Moreover, we develop a new particle filter framework that tracks the human head by using the estimated human body position simultaneously. Our experimental results demonstrate the effectiveness and robustness of the method against several complicated movements of multiple persons. The results also demonstrate that this method can maintain correct tracking when the targets are in close proximity.