UniGaze: Towards Universal Gaze Estimation via Large-scale Pre-Training
This addresses the generalization challenge in gaze estimation for computer vision applications, representing an incremental advance by applying self-supervised pre-training specifically to this domain.
The paper tackles the problem of gaze estimation models failing to generalize across diverse data domains by proposing UniGaze, which uses large-scale self-supervised pre-training on in-the-wild facial datasets, resulting in significant improvements in cross-domain performance while reducing reliance on labeled data.
Despite decades of research on data collection and model architectures, current gaze estimation models encounter significant challenges in generalizing across diverse data domains. Recent advances in self-supervised pre-training have shown remarkable performances in generalization across various vision tasks. However, their effectiveness in gaze estimation remains unexplored. We propose UniGaze, for the first time, leveraging large-scale in-the-wild facial datasets for gaze estimation through self-supervised pre-training. Through systematic investigation, we clarify critical factors that are essential for effective pretraining in gaze estimation. Our experiments reveal that self-supervised approaches designed for semantic tasks fail when applied to gaze estimation, while our carefully designed pre-training pipeline consistently improves cross-domain performance. Through comprehensive experiments of challenging cross-dataset evaluation and novel protocols including leave-one-dataset-out and joint-dataset settings, we demonstrate that UniGaze significantly improves generalization across multiple data domains while minimizing reliance on costly labeled data. source code and model are available at https://github.com/ut-vision/UniGaze.