Fast Tracking of Hands and Fingertips in Infrared Images for Augmented Desk
Yoshinori Kobayashi, Yoichi Sato, Hideki Koike

abstract
In this paper, we introduce a new method for tracking reliably positions of hand center and fingertips of both left and right hands in real-time. Our method makes the use of infrared camera images, and it is based on template matching strategy for finding fingertips. This method composes an essential part of our augmented desk interface in which a user can manipulate both physical objects and electrically projected objects on a desk simultaniously, e.g., a textbook and related WWW pages, with a natural hand gestures. Previous tracking methods which are typically based on color segmentation or background subtraction simply do not perform well in this type of application because an observed color of human skin and image backgrounds may change significantly due to projection of various objects onto a desk. In constrast, our proposed method was shown to be effective even in such a challenging situation through demonstration in our augmented desk interface. This paper explains the details of our tracking method and describes typical applications in our augmented desk interface.