Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation - Télécom Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

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

Tianyu Li
  • Fonction : Auteur
Subhankar Roy
  • Fonction : Auteur
  • PersonId : 1360564
Huayi Zhou
  • Fonction : Auteur
Hongtao Lu
  • Fonction : Auteur

Résumé

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 \textbf{pixel-to-prototype contrast}. Our extensive experiments demonstrate that our method outperforms existing state-of-the-art methods using DeepLabV2. Our code is available at https://github.com/cxa9264/CONFETI

Dates et versions

hal-04205019 , version 1 (12-09-2023)

Identifiants

Citer

Tianyu Li, Subhankar Roy, Huayi Zhou, Hongtao Lu, Stéphane Lathuilière. Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, Vancouver, Canada. ⟨hal-04205019⟩
5 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More