An Intelligent Interactive System Based on Hand Gesture Recognition Algorithm and Kinect

Abstract

In this paper, gesture recognition algorithm with kinect sensor is proposed. the depth cue is used to locate the hand area. Based on the histograms of oriented gradient (HOG) and adaboost learning methods, the static hand algorithm is designed to recognize the predefine gesture in the hand Area. by tracking the hand trajectory by kinect, hmms is used to train and classify dynamic gesture. an intelligent interactive system based on hand gesture recognition algorithm above is presented to choose and control different programs. Experimental results demonstrate that our gesture recognition algorithm is robust and effective, and the built interactive system with more intelligence improves the convenience and nature between human-computer interaction.