Causal Representation-Based Domain Generalization on Gaze Estimation

TL;DR

This paper introduces CauGE, a causal representation framework that improves gaze estimation across different domains by extracting invariant features, achieving state-of-the-art domain generalization performance.

Contribution

The paper proposes a novel causal representation-based approach for domain generalization in gaze estimation, utilizing adversarial training and attention mechanisms to extract invariant features.

Findings

Achieves state-of-the-art domain generalization results.

Effectively extracts domain-invariant features.

Reduces influence of spurious correlations.

Abstract

The availability of extensive datasets containing gaze information for each subject has significantly enhanced gaze estimation accuracy. However, the discrepancy between domains severely affects a model's performance explicitly trained for a particular domain. In this paper, we propose the Causal Representation-Based Domain Generalization on Gaze Estimation (CauGE) framework designed based on the general principle of causal mechanisms, which is consistent with the domain difference. We employ an adversarial training manner and an additional penalizing term to extract domain-invariant features. After extracting features, we position the attention layer to make features sufficient for inferring the actual gaze. By leveraging these modules, CauGE ensures that the neural networks learn from representations that meet the causal mechanisms' general principles. By this, CauGE generalizes across domains by extracting domain-invariant features, and spurious correlations cannot influence the model. Our method achieves state-of-the-art performance in the domain generalization on gaze estimation benchmark.