Real-Time Hand Gesture Recognition for Service Robot

A real-time hand gesture recognition system is developed for human-robot interaction of service robot. The proposed system is mainly composed of two subsystems: one for gesture recognition, and the other for the classification of the gesture motion. The system first uses a cascade classifier to locate the potential hand region from video frame. Then, Gabor wavelets transformation is applied to extract the gesture features which are automatically recognized based on a bank of Support Vector Machines (SVMs). For the estimated motion trajectory of each gesture, we make a set of discrete symbols using vector quantization method and, this symbol sequence is fed into the Hidden Markov Model (HMM) in the gesture motion classification subsystem. Experimental results are shown finally.