Computer Vision for Computer Interaction Computer Vision for Computer Interaction

Abstract

It might seem that an interface based on computer vision would require visual competence near the level of a human being, which is still beyond the state of the art. Fortunately, the interactive application often constrains the vision problem to be solved, allowing fast and simple vision algorithms to be used. This paper gives a brief survey of existing vision-based interactive systems. These systems typically use a number of basic vision algorithms. We describe the basic algorithms used by some systems we have built at MERL: visionbased computer games, a television set controlled by hand gestures, and 3-D head tracking. To appear in: SIGGRAPH Computer Graphics magazine, November, 1999. This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonpro t educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Information Technology Center America; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Information Technology Center America. All rights reserved. Copyright c Mitsubishi Electric Information Technology Center America, 1999 201 Broadway, Cambridge, Massachusetts 02139 MERL, a Mitsubishi Electric Research Lab, 201 Broadway, Cambridge, MA 02139. freeman@merl.com Mitsubishi Electric, Advanced Technology R&D Center, 8-1-1, Tsukaguchi-Honmachi, Amagasaki City, Hyogo 661, Japan present address: E.piphany Software, San Mateo, CA Computer vision for computer interaction W. T. Freeman , P. A. Beardsley , H. Kage, K. Tanaka, K. Kyuma, C. D. Weissman z