Generalizing Eye Tracking With Bayesian Adversarial Learning

• Published in: Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) pp. 11899 - 11908
• Main Authors: Wang, Kang, Zhao, Rui, Su, Hui, Ji, Qiang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2019
• Subjects: Adaptation models; Adversarial machine learning; and Body Pose; Bayes methods; Benchmark testing; Computational modeling; Computer vision; Estimation; Face; Gaze tracking; Gesture; Pattern recognition; Real-time systems
• ISSN: 1063-6919
• DOI: 10.1109/CVPR.2019.01218

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.