Grasp Gesture Recognition Based on Rough Sets Theory in AR System

Abstract

Real-time human-computer interaction is one of the three important features of augmented reality technology. H and i s t he par t of bo dy used m ost frequ ently for th e inte ractions b etween u sers an d objects in virtual scene. Because of the complexity and flexibility of hand motions, it is a hot research topic to recognize the user's grasp gesture accurately and to determine whether the grasp is successful. In this pa per, grasp g esture recogni tion a lgorithm based on roug h sets theo ry wa s proposed a fter analyzing the structure of hand and the features of grasp action, and a related system was developed. The results show that by using this method, users can grasp the virtual objects easily. The interaction process is accurate and natural, which enhances the users' immersion feeling.