Gesture Recognition Using Higher Correlation Feature Information and PCA

Abstract

AbstractThis paper describes the algorithm that lowers the dimension, maintains the gesture recognition and significantly reducesthe eigenspace configuration time by combining the higher correlation feature information and Principle ComponentAnalysis. Since the suggested method doesn't require a lot of computation than the method using existing geometricinformation or stereo image, the fact that it is very suitable for building the real-time system has been proved throughthe experiment. In addition, since the existing point to point method which is a simple distance calculation has manyerrors, in this paper to improve recognition rate the recognition error could be reduced by using several successive inputimages as a unit of recognition with K-Nearest Neighbor which is the improved Class to Class method. Key words : Eigen Space, PCA(Principal Component Analysis), HLAF(Higher order Local Auto Correlation Features) 1. Introduction Men provide various information using non-verbalmeans such as gestures and facial expressions. If it ispossible to analyze these non-verbal communicationmeans, it would be possible to build a natural and intel-ligent interface between man and computer. For analyz-ing human motions, existing 2D systems have manylimitations in motion