Accurate Regression-Based 3D Gaze Estimation Using Multiple Mapping Surfaces

Abstract

Accurate 3D gaze estimation using a simple setup remains a challenging issue for headmounted eye tracking. Current regression-based gaze direction estimation methods implicitly assume that all gaze directions intersect at one point called the eyeball pseudo-center. The effect of this implicit assumption on gaze estimation is unknown. In this paper, we find that this assumption is approximate based on a simulation of all intersections of gaze directions, and it is conditional based on a sensitivity analysis of the assumption in gaze estimation. Hence, we propose a gaze direction estimation method with one mapping surface that satisfies conditions of the assumption by configuring one mapping surface and achieving a highquality calibration of the eyeball pseudo-center. This method only adds two additional calibration points outside the mapping surface. Furthermore, replacing the eyeball pseudo-center with an additional calibrated surface, we propose a gaze direction estimation method with two mapping surfaces that further improves the accuracy of gaze estimation. This method improves accuracy on the state-of-the-art method by 20 percent (from a mean error of 1.84 degrees to 1.48 degrees) on a public dataset with a usage range of 1 meter and by 17 percent (from a mean error of 2.22 degrees to 1.85 degrees) on a public dataset with a usage range of 2 meters.

Gaze Estimation Method Combining Facial Feature Extractor with Pyramid Squeeze Attention Mechanism

“…Zhao et al proposed a monocular gaze estimation network utilizing mixed attention to predict gaze points from monocular features and their location information [21]. Wan et al introduced a technique for estimating gaze direction utilizing one mapping surface [22], which accomplished eye-center calibration through the mapping surface, simplifying the model by making assumptions. Zhuang et al developed a simplified network model for gaze estimation based on the LeNet neural network, called SLeNet [23], utilizing depth-separable convolution to decrease the number of convolution parameters and improve the speed of the model.…”

High-Accuracy 3D Gaze Estimation with Efficient Recalibration for Head-Mounted Gaze Tracking Systems

“…The visual axis is determined by the fixed eyeball center and the estimated gaze point on the screen. As an improved method, the method in [19] requires two additional calibration points outside the mapping surface, then a more precise position of the eyeball center is calculated by triangulation. In [9], the calibration data are collected by staring at a fixed point while rotating head, the position of the eyeball center is set to an estimated initial value, and the loss function based on the angular error of the visual axis is employed to optimize parameters.…”

“…The first one is how to formulate an appropriate regression model. Most paradigms utilize the image pupil center and the gaze point as input and output features [9,19]. However, it may lead to inadequate fitting performance and appreciable extrapolation errors due to the complexity of the human visual system.…”

“…The method based on mapping surfaces [19] mapped the eye image feature to 3D gaze point on a certain plane. This way, two calibration surfaces with different depths correspond to two different regression mapping functions.…”