A Generalized Label Shift Perspective for Cross-Domain Gaze Estimation
This work addresses domain adaptation for gaze estimation, an incremental improvement for real-world applications like human-computer interaction.
The paper tackles cross-domain gaze estimation by introducing a Generalized Label Shift perspective to address domain shift, proposing a correction framework with importance reweighting and conditional operator discrepancy estimation, which shows superior generalization in experiments.
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.