Research of hand positioning and gesture recognition based on binocular vision

Abstract

This paper proposes a new method of extracting feature points of hand. The method uses the center of mass of the hand as the match point to calculate the location information of the target based on Mathematical model of binocular visual positioning. The convex hull points of hand contour obtained by image segmentation can be used to identify the different gestures. Furthermore, a system with both functions of locating the three-dimensional position of hand and identifying the appropriate gestures is designed, which can serve as the interface to drive virtual hand to complete manipulation of grasping, moving and releasing virtual objects.