DYNAMIC HAND GESTURE RECOGNITION USING HIERARCHICAL DYNAMIC BAYESIAN NETWORKS THROUGH LOW-LEVEL IMAGE PROCESSING
Dynamic gesture recognition in video stream has been studied extensively in recent years. To provide efficient and consistent of dynamic hand gesture recognition technique, Hierarchical Dynamic Vision Model (HDVM) which based on dynamic Bayesian networks (DBNs) is proposed for automatically recognizing human hand gestures in this paper. HDVM consists of the fast differential color tracking algorithm (DCTA) for tracking object trajectory and the motion pattern analyzer (MPA) for recognizing the hand gestures. In this paper, the proposed model is able to recognize three dynamic hand gestures through the low-level image analysis. In the low-level image processing, both motion trajectories and motion directions generated from hand part are used as features after segmentation.
Dynamic Bayesian networks Fast differential color tracking algorithm Dynamic hand gesture recognition
WEI-HUA ANDREW WANG CHUN-LIANG TUNG
Department of Industrial Engineering and Enterprise Information, Tunghai University, Taichung, Taiwan
国际会议
2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)
昆明
英文
3247-3253
2008-07-12（万方平台首次上网日期，不代表论文的发表时间）