I n this paper, we propose a method for controlling mouse pointers o n unknown computer displays by the face motions of computer users viewed f r o m uncalibrated cameras.Especially, we show that by considering a f i n e bases o n computer displays, we can derive one-to-one correspondences between gaze positions o n displays and the projections of face features in images.T h e proposed method is implemented and tested in real t i m e experiments.
SUMMARYA contact-type interface (such as a mouse or a keyboard) is usually used on personal computers. However, this is not convenient if the operator has to use his or her hands for another operation, or if his or her hands are disabled. To deal with such cases, this paper proposes a visual interface in which the mouse pointer on the computer display is controlled by the movement of the operator's eyes via a camera image. Conventional visual interfaces require tedious calibrations of the camera and system. The proposed visual interface works accurately without any calibration. In the proposed system, the gaze point on the display screen can be reliably calculated from the user's projected image by using projective bases and affine bases and the movement of the user's viewing direction. Control of the mouse pointer by means of the proposed system has been successfully tested. The usefulness of the proposed method has been confirmed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.