Contrastive Representation Learning for Hand Shape Estimation
Contrastive Representation Learning for Hand Shape Estimation
复制标题
手形估计的对比表示学习
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
T. Brox
中科院分区：
文献类型：
--
作者：
Christiane Zimmermann;Max Argus;T. Brox
这项工作基于无监督学习的最新进展，提出了单眼手形估计的改进。我们扩展了动量对比学习，并贡献了一个结构化的手部图像集合，非常适合视觉表征学习，我们称之为 HanCo。我们发现，通过利用先进的背景去除技术和多视图信息，可以显着改善通过已建立的对比学习方法学习的表示。这些使我们能够生成比基于示例的方法中常用的增强所获得的实例对更加多样化的实例对。与 ImageNet 预训练基线相比，我们的方法为手形估计任务提供了更合适的表示，并且网格误差减少了 4.7%，F 分数提高了 3.6%。我们公开我们的基准数据集，以鼓励进一步研究这个方向。
This work presents improvements in monocular hand shape estimation by building on top of recent advances in unsupervised learning. We extend momentum contrastive learning and contribute a structured collection of hand images, well suited for visual representation learning, which we call HanCo. We find that the representation learned by established contrastive learning methods can be improved significantly by exploiting advanced background removal techniques and multi-view information. These allow us to generate more diverse instance pairs than those obtained by augmentations commonly used in exemplar based approaches. Our method leads to a more suitable representation for the hand shape estimation task and shows a 4.7% reduction in mesh error and a 3.6% improvement in F-score compared to an ImageNet pretrained baseline. We make our benchmark dataset publicly available, to encourage further research into this direction.