New algorithm for determining 3D coordinates of the eye cornea center in a non -contact eye-tracking system

In a number of non-contact systems for tracking the direction of gaze, based on video data generated by a video camera remote from the client's head, the relative position of the camera and the client's eyes is determined by the glares created on the cornea of the eyes by point sources of IR radiation rigidly connected to the camera. We propose in this paper a new algorithm for estimating the three-dimensional coordinates of the cornea center from images of glares generated by a pair of point radiation sources. We show that when the distance between the eye and the camera is much greater than the radius of the cornea, the problem comes down to solving one nonlinear equation, which significantly reduces computational costs. The results of a numerical experiment confirm the preservation of the estimates accuracy.