Depth and Skeleton Information Model for Kinect Based Hand Segmentation

Abstract

With the rapid development of the lowcost Microsoft Kinect, hand segmentation has been a resurgence of broad interest. This is because the depth and skeleton information provided by the Kinect opens an innovative way for hand segmentation. In this paper, we propose a new scheme for hand detection and segmentation based on depth and skeleton information. We conduct experiments on our new collect RGB and depth image pairs. The results demonstrate the robustness and effectiveness of our proposed model.