In-domain representation learning for remote sensing

TL;DR

This paper introduces a standardized approach to in-domain representation learning for remote sensing, providing baselines and evaluating dataset characteristics to improve generic remote sensing representations.

Contribution

It offers a unified framework and baseline results for remote sensing representation learning, addressing a gap in the community and establishing evaluation protocols.

Findings

Established state-of-the-art performance on multiple datasets

Provided simplified access to diverse remote sensing datasets

Analyzed dataset characteristics important for effective representation learning

Abstract

Given the importance of remote sensing, surprisingly little attention has been paid to it by the representation learning community. To address it and to establish baselines and a common evaluation protocol in this domain, we provide simplified access to 5 diverse remote sensing datasets in a standardized form. Specifically, we investigate in-domain representation learning to develop generic remote sensing representations and explore which characteristics are important for a dataset to be a good source for remote sensing representation learning. The established baselines achieve state-of-the-art performance on these datasets.