Head pose-free appearance-based gaze sensing via eye image synthesis

This paper addresses the problem of estimating hu-man gaze from eye appearance under free head mo-tion. Allowing head motion remains challenging be-cause eye appearance changes significantly for differ-ent head poses, and thus new head poses require new training images. To avoid repetitive training, we pro-pose to produce synthetic training images for varying head poses. First, we model pixel displacements be-tween head-moving eye images as 1D pixel flows, and then produce such flows to synthesize new training im-ages from the original training images captured under a fixed default head pose. Specifically, we produce all the required 1D flows by using only four additionally cap-tured images. Our method was successfully tested with extensive experiments to demonstrate its effectiveness. 1