Abstract
Gesture recognition is a kind of interaction of the natural user interface (Natural User Interface), which plays an important role in the application of the body-experience. In this paper, it uses Kinect sensor to track the 20 skeleton joint points of human body, and calculates the distances and angles between the nodes to determine the specific body posture, and then responses to a specific function which is set up and gets somatosensory control effect. Experiments show that, this method has higher recognition rate, short response time, and has a good scalability. The algorithm can meet the common body sense experience requirement of the virtual environment, and improves the efficiency of using virtual.
Publisher
Trans Tech Publications, Ltd.
Reference5 articles.
1. WU X, LUO Y. Natural Interaction Design on the Internet of Things (In Chinese) [J]. DESIGN RESEARCH, 2013, 3(5): 5-8.
2. Webb J, Ashley J. Beginning Kinect Programming with the Microsoft Kinect SDK [M]. [s. l. ]: Apress, (2012).
3. XIE L, LIAO H, YANG Y. Recognition and Application Research of Kinect-based Gesture (In Chinese) [J]. COMPUTER TECHNOLOGY AND DEVELOPMENT, 2013, 23(5): 258-260.
4. YU L, WANG H, ZHONG H. Face Recognition Based on Hidden Markov Model (In Chinese) [J]. COMPUTER TECHNOLOGY AND DEVELOPMENT, 2012, 22(2): 25-28.
5. Chu T, Su CH. A Kinect-based Handwritten Digit Recognition for TV Remote Controller [C] . 2012 IEEE International Symposiums on Intelligent Signal Processing and Communication Systems (ISPACS 2012), 2012, November 4-7: 414-419.
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献