Research of hand gesture using Kinect based on finger recognition

Zhuo Shi^{1, 2, 3}, Yinghui Li², Ke Yu², Yuanquan Cheng², Changshao Zhou² 

¹South China University of Technology, Guangzhou, Guangdong 510275, China

²Computer Science & Engineering of the Guilin University of Electronic Technology, Guilin, Guangxi 541004, China

³Guangxi Key Laboratory of Trusted Software (Guilin University of Electronic Technology)

Using the depth image from Kinect, we present a novel method to archive the Hand Gesture Recognition (HGR). Firstly we overview the major technical components of the complete method. Then we elaborate several key challenges such as palm recognition, contour analysis and fingertip detection. We present the Biggest concave point detect schemes to contour analysis which is the key to achieving the HGR. Finally, we demonstrate the feasibility and effectiveness of the method by running the system.