A 3D gaze estimation method based on facial feature tracking

Abstract

A 3D gaze estimation and tracking algorithm based on facial feature tracking is present in this paper. By using two cameras, 2D facial feature points are detected and tracked with Active Shape Models (ASM). Then, the 3D coordinates of the feature points on a face can be obtained by stereo vision. The full 3D pose of head is obtained by comparing the feature points of current pose with initial head pose. The head pose estimation eliminate the restraint of head movement. Based on a 3D eye model with facial feature points, 3D visual axis can be obtained by estimating the 3D pupil center and head pose after a one-time personal calibration. The experimental results show the accuracy of our gaze tracking system achieves less than 3 degree.