Appearance-based gaze estimation using kinect
Jinsoo Choi, Byungtae Ahn, Jaesik Parl, In So Kweon · 2013
Human gaze tracking has gathered much attention due to its capability to detect intuitive attention. Appearance-based methods can work with a single camera in ordinary conditions to track human gaze. An effective way to generate eye appearances is proposed using the Kinect. The head pose information is obtained from the Kinect after a series of calibrations. The Eye Appearance features are collected through ASM and KLT feature tracker. With 23 training samples, the error is found to be 1.07°. This paper proposes an efficient scheme for gaze tracking using a single Kinect device.