Deep domain adaptation in earth observation

Benjamin Kellenberger*, Onur Tasar, Bharath Bhushan Damodaran, Nicolas Courty, Devis Tuia

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

7 Citations (Scopus)

Abstract

When applied to new datasets, acquired at different time moments, with different sensors or under different acquisition conditions, deep learning models might fail spectacularly. This is because they have learned from the data distribution observed during training and, as such, do not generalize out of that domain naturally. This chapter introduces methodologies designed to tackle this problem and provide deep learning models able to adapt to new data distributions, i.e. domain adaptation. Domain adaptation works by either adapting the representation to the new data distribution, modifying the inputs or performing smart sampling. But independently of the strategy, they lead to updated models, able to process effectively the new data without needing observation from it (or a very limited amount).

Original languageEnglish
Title of host publicationDeep Learning for the Earth Sciences
Subtitle of host publicationA Comprehensive Approach to Remote Sensing, Climate Science and Geosciences
EditorsGustau Camps-Valls, Devis Tuia, Xiao Xiang Zhu, Markus Reichstein
PublisherWiley
Chapter7
Pages90-104
Number of pages15
ISBN (Electronic)9781119646181
ISBN (Print)9781119646143
DOIs
Publication statusPublished - 20 Aug 2021

Keywords

  • Data distribution
  • Deep learning
  • Domain adaptation
  • Earth observation
  • Machine learning

Fingerprint

Dive into the research topics of 'Deep domain adaptation in earth observation'. Together they form a unique fingerprint.

Cite this