StandardGAN: Multi-source Domain Adaptation for Semantic Segmentation of Very High Resolution Satellite Images by Data Standardization - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

StandardGAN: Multi-source Domain Adaptation for Semantic Segmentation of Very High Resolution Satellite Images by Data Standardization

Résumé

Domain adaptation for semantic segmentation has recently been actively studied to increase the generalization capabilities of deep learning models. The vast majority of the domain adaptation methods tackle single-source case, where the model trained on a single source domain is adapted to a target domain. However, these methods have limited practical real world applications, since usually one has multiple source domains with different data distributions. In this work, we deal with the multi-source domain adaptation problem. Our method, namely StandardGAN, standardizes each source and target domains so that all the data have similar data distributions. We then use the standardized source domains to train a classifier and segment the standardized target domain. We conduct extensive experiments on two remote sensing data sets, in which the first one consists of multiple cities from a single country, and the other one contains multiple cities from different countries. Our experimental results show that the standardized data generated by StandardGAN allow the classifiers to generate significantly better segmentation.
Fichier principal
Vignette du fichier
standardgan.pdf (8.75 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02893537 , version 1 (08-07-2020)

Identifiants

  • HAL Id : hal-02893537 , version 1

Citer

Onur Tasar, Yuliya Tarabalka, Alain Giros, Pierre Alliez, Sébastien Clerc. StandardGAN: Multi-source Domain Adaptation for Semantic Segmentation of Very High Resolution Satellite Images by Data Standardization. CVPR 2020 - IEEE/CVF Conference on Computer Vision and Pattern Recognition EarthVision Workshop, Jun 2020, Seattle / Virtual, United States. ⟨hal-02893537⟩
98 Consultations
35 Téléchargements

Partager

Gmail Facebook X LinkedIn More