Generalizing Eye Tracking with Bayesian Adversarial Learning

Existing appearance-based gaze estimation approaches with CNN have poor generalization performance. By systematically studying this issue, we identify three major factors: 1) appearance variations; 2) head pose variations and 3) over-fitting issue with point estimation. To improve the generalization performance, we propose to incorporate adversarial learning and Bayesian inference into a unified framework. In particular, we first add an adversarial component into traditional CNN-based gaze estimator so that we can learn features that are gaze-responsive but can generalize to appearance and pose variations. Next, we extend the point-estimation based deterministic model to a Bayesian framework so that gaze estimation can be performed using all parameters instead of only one set of parameters. Besides improved performance on several benchmark datasets, the proposed method also enables online adaptation of the model to new subjects/environments, demonstrating the potential usage for practical real-time eye tracking applications.

Reference

Kang Wang, Rui Zhao, Hui Su, Qiang Ji, "Generalizing Eye Tracking with Bayesian Adversarial Learning,"

Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 11907-11916

Bibtex

@InProceedings{Wang_2019_CVPR,
author = {Wang, Kang and Zhao, Rui and Su, Hui and Ji, Qiang},
title = {Generalizing Eye Tracking With Bayesian Adversarial Learning},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}