Gaze Estimation Based on 3D Face Structure and Pupil Centers

Abstract

It is a challenging problem to realize a robust and low cost gaze estimation system. Most existing feature-based gaze estimation methods strongly rely on cornea reflections, which are unstable to glasses, head movements and natural light. In this paper, we propose a novel gaze estimation method without use of cornea reflections based on a stereo camera system. Firstly, 3D Active Shape Models (ASM) is reconstructed using stereo vision to represent 3D face structure. Then, without use of cornea reflections, a 3D Pupil-Eye-Contours based feature is proposed to represent human gaze information. What's more, precise estimation of head poses based on 3D face structure is employed to rectify the 3D pupil centers and eye contours for improving the ability of tolerance to head movements. Experiments on fifteen subjects show that the system is accurate and allows natural head movements.