A Generalized Label Shift Perspective for Cross-Domain Gaze Estimation

TL;DR

This paper introduces a novel generalized label shift perspective for cross-domain gaze estimation, proposing a correction framework with importance reweighting to improve model generalization across different domains.

Contribution

It models cross-domain gaze estimation as a label and conditional shift problem and presents a GLS correction framework with importance reweighting for better domain adaptation.

Findings

Outperforms existing methods in cross-domain gaze estimation tasks.

Demonstrates superior generalization across various backbone models.

Validates effectiveness through extensive experiments.

Abstract

Aiming to generalize the well-trained gaze estimation model to new target domains, Cross-domain Gaze Estimation (CDGE) is developed for real-world application scenarios. Existing CDGE methods typically extract the domain-invariant features to mitigate domain shift in feature space, which is proved insufficient by Generalized Label Shift (GLS) theory. In this paper, we introduce a novel GLS perspective to CDGE and modelize the cross-domain problem by label and conditional shift problem. A GLS correction framework is presented and a feasible realization is proposed, in which a importance reweighting strategy based on truncated Gaussian distribution is introduced to overcome the continuity challenges in label shift correction. To embed the reweighted source distribution to conditional invariant learning, we further derive a probability-aware estimation of conditional operator discrepancy. Extensive experiments on standard CDGE tasks with different backbone models validate the superior generalization capability across domain and applicability on various models of proposed method.