Gesture based improved human-computer interaction using Microsoft's Kinect sensor

Abstract

A simple and robust gesture recognition system is proposed for better human-computer interaction using Microsoft's Kinect sensor. The Kinect is employed to construct skeletons for a subject in the 3D space using twenty body joint coordinates. From this skeletal information, ten joints are required and six triangles have been constructed along with six respective centroids. The feature space corresponds to the Euclidean distances between spine joint and the centroids for each frame. For classification purpose, support vector machine is used using a kernel function. The proposed work is widely applicable for several gesture driven computer applications and produces an average accuracy rate of 88.7%.