OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Mapping

TL;DR

OpenEarthMap is a comprehensive high-resolution land cover dataset covering 97 regions globally, enabling the development and evaluation of semantic segmentation models for universal land mapping applications.

Contribution

It introduces a large-scale, annotated dataset for global land cover mapping and evaluates models' generalization, domain adaptation, and lightweight architectures.

Findings

Models trained on OpenEarthMap generalize well worldwide.

Unsupervised domain adaptation improves model performance across regions.

Lightweight models via neural architecture search enable fast, resource-efficient mapping.

Abstract

We introduce OpenEarthMap, a benchmark dataset, for global high-resolution land cover mapping. OpenEarthMap consists of 2.2 million segments of 5000 aerial and satellite images covering 97 regions from 44 countries across 6 continents, with manually annotated 8-class land cover labels at a 0.25--0.5m ground sampling distance. Semantic segmentation models trained on the OpenEarthMap generalize worldwide and can be used as off-the-shelf models in a variety of applications. We evaluate the performance of state-of-the-art methods for unsupervised domain adaptation and present challenging problem settings suitable for further technical development. We also investigate lightweight models using automated neural architecture search for limited computational resources and fast mapping. The dataset is available at https://open-earth-map.org.