Robust hand detection using arm segmentation from depth data and static palm gesture recognition

Abstract

Nowadays there is a big interest in applying natural gesture communication language to various systems. There are several techniques on how to detect willed hand motion or recognize gestures by using different kind of sensors. RGB cameras can be used to detect hand, but it has a limited application. For example, hand detection can be particularly hard in some lighting conditions or in different skin colors. Using depth camera data, it is possible to distinguish object in front of camera in a better way. In this paper, we present a technique to detect active user and his hand gesture using both RGB and depth data. The approach firstly detects the user's face, mapping it to depth map data and finally uses the region from camera to person for human arm detection and hand palm recognition.