An Integrated Gesture Recognition Scheme for Human Computer Interactions

Vision-based hand gesture recognition is a popular research topic for Human-Machine Interaction (HMI). Hand gestures having both local and global motions are most common form of gestures. Examples of such gestures are signs used in different sign languages. Recognition of these gestures require both the hand trajectory and hand shape information. The proposed system is designed so that it can distinguish and recognize different classes of hand gestures having different spatio-temporal behaviour. The system uses finite state machine, where hand shape information along with temporal characteristics of the gesticulating hand plays important role in recognition. Moreover the spatial hand position and hand motion features are considered for recognition. Experimental results show that the proposed system is very much suitable for the applications related to HCI.