A hand tracking framework using the 3D active tracking volume

Abstract

In this paper, we present a hand tracking and gesture recognition system. Our system employs a depth capture device to acquire 3D geometric information of user's bare hand, and the user's hand position and gesture information is estimated from the depth clues. In particular, we build a flexible tracking volume and restrict the hand tracking area within the tracking volume in order to avoid diverse problems caused by conventional object detection/ tracking systems. This tracking area is actively adjusted according to the statistics of user's hand gesture and motion history. The proposed system aims to be adopted to be employed in diverse systems which are operated on the basis of natural user interface framework. The experimental results verify accuracy and efficiency of the proposed system.