Cross-Region Domain Adaptation for Class-level Alignment

14 Sep 2021  ·  Zhijie Wang, Xing Liu, Masanori Suganuma, Takayuki Okatani ·

Semantic segmentation requires a lot of training data, which necessitates costly annotation. There have been many studies on unsupervised domain adaptation (UDA) from one domain to another, e.g., from computer graphics to real images. However, there is still a gap in accuracy between UDA and supervised training on native domain data. It is arguably attributable to class-level misalignment between the source and target domain data. To cope with this, we propose a method that applies adversarial training to align two feature distributions in the target domain. It uses a self-training framework to split the image into two regions (i.e., trusted and untrusted), which form two distributions to align in the feature space. We term this approach cross-region adaptation (CRA) to distinguish from the previous methods of aligning different domain distributions, which we call cross-domain adaptation (CDA). CRA can be applied after any CDA method. Experimental results show that this always improves the accuracy of the combined CDA method, having updated the state-of-the-art.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Domain Adaptation GTA5 to Cityscapes ProDA+CRA mIoU 58.6 # 18
Synthetic-to-Real Translation GTAV-to-Cityscapes Labels ProDA + CRA mIoU 58.6 # 21
Semantic Segmentation GTAV-to-Cityscapes Labels ProDA+CRA mIoU 58.6 # 9
Image-to-Image Translation GTAV-to-Cityscapes Labels ProDA+CRA mIoU 58.6 # 14
Domain Adaptation Synscapes-to-Cityscapes ProDA+CRA mIoU 60.2 # 1
Synthetic-to-Real Translation SYNTHIA-to-Cityscapes ProDA+CRA MIoU (13 classes) 63.7 # 13
MIoU (16 classes) 56.9 # 13
Image-to-Image Translation SYNTHIA-to-Cityscapes ProDA+CRA mIoU (13 classes) 63.7 # 10
Domain Adaptation SYNTHIA-to-Cityscapes ProDA+CRA mIoU 56.9 # 14

Methods


No methods listed for this paper. Add relevant methods here