See Through the Noise: Improving Domain Generalization in Gaze Estimation

TL;DR

This paper introduces SeeTN, a framework that enhances domain generalization in gaze estimation by mitigating label noise through semantic embedding and affinity regularization.

Contribution

It is the first to systematically analyze label noise effects in gaze estimation and proposes a novel noise mitigation method that improves cross-domain robustness.

Findings

SeeTN effectively reduces the impact of label noise on model performance.

The framework improves cross-domain generalization without losing source accuracy.

Extensive experiments validate the superiority of SeeTN over existing methods.

Abstract

Generalizable gaze estimation methods have garnered increasing attention due to their critical importance in real-world applications and have achieved significant progress. However, they often overlook the effect of label noise, arising from the inherent difficulty of acquiring precise gaze annotations, on model generalization performance. In this paper, we are the first to comprehensively investigate the negative effects of label noise on generalization in gaze estimation. Further, we propose a novel solution, called See-Through-Noise (SeeTN) framework, which improves generalization from a novel perspective of mitigating label noise. Specifically, we propose to construct a semantic embedding space via a prototype-based transformation to preserve a consistent topological structure between gaze features and continuous labels. We then measure feature-label affinity consistency to distinguish noisy from clean samples, and introduce a novel affinity regularization in the semantic manifold to transfer gaze-related information from clean to noisy samples. Our proposed SeeTN promotes semantic structure alignment and enforces domain-invariant gaze relationships, thereby enhancing robustness against label noise. Extensive experiments demonstrate that our SeeTN effectively mitigates the adverse impact of source-domain noise, leading to superior cross-domain generalization without compromising the source-domain accuracy, and highlight the importance of explicitly handling noise in generalized gaze estimation.