Real-time gesture recognition using Maskable Template Model

Abstract

We describe a new method for real-time gesture recognition and its application. The key idea of the method is that we use multiple template models for specific feature parts of the body such as arms and estimate the pose of this part by comparing similarities of an input image to the models. To calculate the similarity for the part, we developed the Maskable Template Model (MTM) with which both masking and template matching operations are simultaneously completed. As an application of this method, we realize a real-time interactive system, Gesture Game System, which can activate the characters of a video game by gestures in real-time.