Hand gesture segmentation, recognition and application

Abstract

Vision-based hand gesture recognition provides a more nature and powerful means for human-computer interaction. In this paper, we present some new approaches for hand image segmentation, 2D hand posture recognition and 3D hand gesture reconstruction. We segment hand images from complex backgrounds with the color segmentation approach based on the RCE neural network, and recognize 2D hand postures by analyzing the topological features of the segmented hand. We also propose to use the 3D reconstruction techniques to recover 3D hand gestures and present a new method to estimate the fundamental matrix between two uncalibrated cameras from stereo hand images. A gesture-based human-robot interface has been developed to demonstrate the application of hand gestures.