Deep neural networks have significantly improved appearance-based gaze estimation accuracy. However, it still suffers from unsatisfactory performance when generalizing the trained model to new domains, e.g., unseen environments or persons. In this paper, we propose a plug-and-play gaze adaptation framework (PnP-GA), which is an ensemble of networks that learn collaboratively with the guidance of outliers. Since our proposed framework does not require ground-truth labels in the target domain, the existing gaze estimation networks can be directly plugged into PnP-GA and generalize the algorithms to new domains. We test PnP-GA on four gaze domain adaptation tasks, ETH-to-MPII, ETH-to-EyeDiap, Gaze360-to-MPII, and Gaze360-to-EyeDiap. The experimental results demonstrate that the PnP-GA framework achieves considerable performance improvements of 36.9%, 31.6%, 19.4%, and 11.8% over the baseline system. The proposed framework also outperforms the state-of-the-art domain adaptation approaches on gaze domain adaptation tasks.
@article{arxiv.2107.13780,
title = {Generalizing Gaze Estimation with Outlier-guided Collaborative Adaptation},
author = {Yunfei Liu and Ruicong Liu and Haofei Wang and Feng Lu},
journal= {arXiv preprint arXiv:2107.13780},
year = {2021}
}
Comments
This paper is accepted by ICCV2021. Code has been released at https://github.com/DreamtaleCore/PnP-GA