User segmentation using moving edge detection for sports simulator

This paper proposes user segmentation algorithm using moving edge detection to provide customized coaching service to users in advanced sports simulator like horse riding. To give this service in the simulator, exact user segmentation is very important. If the segmenting user from its background became better, we will get a better pose recognition results. In this paper, some of the well-known object segmentation methods to be applied to sports simulator have been investigated. Also, a moving object detection algorithm for recognizing user´s poses which is a novel innovation to assure high recognition reliability in sports simulator has been proposed.