Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation

Tianyu Li, Subhankar Roy, Huayi Zhou, Hongtao Lu, Stéphane Lathuilière; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, pp. 4869-4879

Abstract


To overcome the domain gap between synthetic and real-world datasets, unsupervised domain adaptation methods have been proposed for semantic segmentation. Majority of the previous approaches have attempted to reduce the gap either at the pixel or feature level, disregarding the fact that the two components interact positively. To address this, we present CONtrastive FEaTure and pIxel alignment (CONFETI) for bridging the domain gap at both the pixel and feature levels using a unique contrastive formulation. We introduce well-estimated prototypes by including category-wise cross-domain information to link the two alignments: the pixel-level alignment is achieved using the jointly trained style transfer module with the prototypical semantic consistency, while the feature-level alignment is enforced to cross-domain features with the pixel-to-prototype contrast. Our extensive experiments demonstrate that our method outperforms existing state-of-the-art methods using DeepLabV2. Our code has been made publicly available.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Li_2023_CVPR, author = {Li, Tianyu and Roy, Subhankar and Zhou, Huayi and Lu, Hongtao and Lathuili\`ere, St\'ephane}, title = {Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4869-4879} }