DAugNet: Unsupervised, Multisource, Multitarget, and Life-Long Domain Adaptation for Semantic Segmentation of Satellite Images

• Published in: IEEE transactions on geoscience and remote sensing Vol. 59; no. 2; pp. 1067 - 1081
• Main Authors: Tasar, Onur, Giros, Alain, Tarabalka, Yuliya, Alliez, Pierre, Clerc, Sebastien
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptation; Adaptation models; Classifiers; Convolutional neural networks (CNNs); Data; dense labeling; domain adaptation; Domains; generative adversarial networks (GANs); Geographical locations; Image processing; Image segmentation; Iterative methods; Learning algorithms; life-long adaption; Machine learning; multisource adaption; multitarget adaption; Proliferation; Remote sensing; Satellite imagery; Satellites; Semantic segmentation; Semantics; Spaceborne remote sensing; Standardization; Training
• DOI: 10.1109/TGRS.2020.3006161
• ISSN: 0196-2892

The domain adaptation of satellite images has recently gained increasing attention to overcome the limited generalization abilities of machine learning models when segmenting large-scale satellite images. Most of the existing approaches seek for adapting the model from one domain to another. However, such single-source and single-target setting prevents the methods from being scalable solutions since, nowadays, multiple sources and target domains having different data distributions are usually available. Besides, the continuous proliferation of satellite images necessitates the classifiers to adapt to continuously increasing data. We propose a novel approach, coined DAugNet, for unsupervised, multisource, multitarget, and life-long domain adaptation of satellite images. It consists of a classifier and a data augmentor. The data augmentor, which is a shallow network, is able to perform style transfer between multiple satellite images in an unsupervised manner, even when new data are added over time. In each training iteration, it provides the classifier with diversified data, which makes the classifier robust to large data distribution difference between the domains. Our extensive experiments prove that DAugNet significantly better generalizes to new geographic locations than the existing approaches.