Shallow Features Guide Unsupervised Domain Adaptation for Semantic Segmentation at Class Boundaries
This work addresses the challenge of obtaining sharp segmentation masks in unsupervised domain adaptation for semantic segmentation, which is important for applications like autonomous driving, but it appears incremental as it builds on existing adaptation frameworks.
The paper tackles the problem of domain shift in semantic segmentation, particularly at class boundaries, by introducing a low-level adaptation strategy and a data augmentation method for self-training, resulting in improved sharpness of segmentation masks.
Although deep neural networks have achieved remarkable results for the task of semantic segmentation, they usually fail to generalize towards new domains, especially when performing synthetic-to-real adaptation. Such domain shift is particularly noticeable along class boundaries, invalidating one of the main goals of semantic segmentation that consists in obtaining sharp segmentation masks. In this work, we specifically address this core problem in the context of Unsupervised Domain Adaptation and present a novel low-level adaptation strategy that allows us to obtain sharp predictions. Moreover, inspired by recent self-training techniques, we introduce an effective data augmentation that alleviates the noise typically present at semantic boundaries when employing pseudo-labels for self-training. Our contributions can be easily integrated into other popular adaptation frameworks, and extensive experiments show that they effectively improve performance along class boundaries.