Abstract
In video-based human gesture recognition, it is very important to combine useful features and analyze the dynamic structure thereof as efficiently as possible. In this paper, we proposed a dynamic Bayesian network model that is a simplified model of dynamics at the level of hidden variables and employs observation windows of observation time slices for robust modeling and handling of noise and other variabilities. The proposed Simplified dynamic Bayesian network (DBN) was tested on a gesture database and an American sign language database. According to the experiments, the proposed DBN outperformed other methods: Conditional Random Fields (CRFs), conventional Bayesian Networks (BNs), DBNs, and Hidden Markov Models (HMMs). The proposed DBN achieved 98 % recognition accuracy in gesture recognition and 94.6 % in ASL recognition whereas the HMM and the CRF did 80 and 86 % in gesture recognition and 75.4 and 85.4 % in ASL (American Sign Language) recognition, respectively.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
