Domain Adaptive Semantic Segmentation Using Weak Labels
Domain Adaptive Semantic Segmentation Using Weak Labels
复制标题
DOI：
10.1007/978-3-030-58545-7_33
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
S. Paul;Yi-Hsuan Tsai;S. Schulter;A. Roy-Chowdhury;Manmohan Chandraker
中科院分区：
文献类型：
--
作者：
S. Paul;Yi-Hsuan Tsai;S. Schulter;A. Roy-Chowdhury;Manmohan Chandraker
学习语义分割模型需要大量的像素级标记。然而，带标签的数据可能仅在与所需目标域不同的域中大量可用，该目标域只具有最少的注释或没有注释。在这项工作中，我们提出了一个新的框架，在目标领域的图像级弱标签的语义分割中的领域自适应。弱标签可以基于无监督领域适应(UDA)的模型预测，或者从用于语义分割的新的弱监督领域适应(WDA)范例中的人类注释器获得。使用弱标签既实用又有用，因为(I)在WDA中收集图像级目标注释相对便宜，并且在UDA中不产生成本，以及(Ii)它打开了按类别进行域对齐的机会。我们的框架使用弱标签来实现特征对齐和伪标注之间的相互作用，在领域自适应的过程中都得到了改进。具体地说，我们开发了一个弱标签分类模块来强制网络关注某些类别，然后使用这些训练信号来指导所提出的类别对齐方法。在实验中，我们显示出相对于UDA中现有的最先进水平的相当大的改进，并在WDA设置中提出了一个新的基准。项目页面位于http://www.nec-labs.com/~mas/WeakSegDA。
Learning semantic segmentation models requires a huge amount of pixel-wise labeling. However, labeled data may only be available abundantly in a domain different from the desired target domain, which only has minimal or no annotations. In this work, we propose a novel framework for domain adaptation in semantic segmentation with image-level weak labels in the target domain. The weak labels may be obtained based on a model prediction for unsupervised domain adaptation (UDA), or from a human annotator in a new weakly-supervised domain adaptation (WDA) paradigm for semantic segmentation. Using weak labels is both practical and useful, since (i) collecting image-level target annotations is comparably cheap in WDA and incurs no cost in UDA, and (ii) it opens the opportunity for category-wise domain alignment. Our framework uses weak labels to enable the interplay between feature alignment and pseudo-labeling, improving both in the process of domain adaptation. Specifically, we develop a weak-label classification module to enforce the network to attend to certain categories, and then use such training signals to guide the proposed category-wise alignment method. In experiments, we show considerable improvements with respect to the existing state-of-the-arts in UDA and present a new benchmark in the WDA setting. Project page is at http://www.nec-labs.com/~mas/WeakSegDA .