Jindal Swati, Manduchi Roberto
University of California, Santa Cruz, Santa Cruz, CA, 95064, USA.
Proc Mach Learn Res. 2023;210:37-49.
Self-supervised learning (SSL) has become prevalent for learning representations in computer vision. Notably, SSL exploits contrastive learning to encourage visual representations to be invariant under various image transformations. The task of gaze estimation, on the other hand, demands not just invariance to various appearances but also equivariance to the geometric transformations. In this work, we propose a simple contrastive representation learning framework for gaze estimation, named . exploits multi-view data to promote equivariance and relies on selected data augmentation techniques that do not alter gaze directions for invariance learning. Our experiments demonstrate the effectiveness of for several settings of the gaze estimation task. Particularly, our results show that improves the performance of cross-domain gaze estimation and yields as high as 17.2% relative improvement. Moreover, the framework is competitive with state-of-the-art representation learning methods for few-shot evaluation. The code and pre-trained models are available at https://github.com/jswati31/gazeclr.
自监督学习（SSL）在计算机视觉中学习表征方面已变得十分普遍。值得注意的是，SSL利用对比学习来促使视觉表征在各种图像变换下保持不变。另一方面，注视估计任务不仅要求对各种外观具有不变性，还要求对几何变换具有等变性。在这项工作中，我们提出了一种用于注视估计的简单对比表征学习框架，名为 。 利用多视图数据来促进等变性，并依赖于选定的不会改变注视方向以进行不变性学习的数据增强技术。我们的实验证明了 在注视估计任务的几种设置下的有效性。特别是，我们的结果表明 提高了跨域注视估计的性能，相对提升高达17.2%。此外， 框架在少样本评估方面与当前最先进的表征学习方法具有竞争力。代码和预训练模型可在https://github.com/jswati31/gazeclr获取。
