Identifying Body Parts of Multiple People in Multi-Camera Images

In order to track and recognize the movements of multiple people using multiple cameras, each person needs to be segmented and identified in the image of each camera. We propose a method that tracks multiple people and identifies their body parts in multiple camera images. Estimation of the human positions and identification among the multiple cameras is mainly based on the silhouettes method combining the background subtraction and the frame subtraction. The noise problem and the combine problem of the persons in the silhouettes method is addressed by use of the existence probability map. The existence probability map consists of the position probability map and height cumulus map. The gesture event is detected by the human block and motion block that are extracted from the voting space using the silhouettes method. We demonstrate the experimental results and the validity of the proposed method.