license: cc-by-4.0
task_categories:
- image-feature-extraction
tags:
- earth-observation
- remote-sensing
- satellite-imagery
- multimodal
- sentinel
- pretraining-data
size_categories:
- 10M<n<100M
Copernicus-Pretrain Dataset
This repository contains the Copernicus-Pretrain dataset, part of the work presented in the paper Towards a Unified Copernicus Foundation Model for Earth Vision.
The official implementation and more details about the Copernicus Foundation Model project can be found on GitHub: https://github.com/wangyi111/Copernicus-FM.
Introduction
Recent advances in Earth observation (EO) foundation models have unlocked the potential of big satellite data to learn generic representations from space, benefiting a wide range of downstream applications crucial to our planet. This work takes a step towards next-generation EO foundation models with three key components:
- Copernicus-Pretrain: A massive-scale pretraining dataset that integrates 18.7M aligned images from all major Copernicus Sentinel missions, spanning from the Earth's surface to its atmosphere.
- Copernicus-FM: A unified foundation model capable of processing any spectral or non-spectral sensor modality using extended dynamic hypernetworks and flexible metadata encoding.
- Copernicus-Bench: A systematic evaluation benchmark with 15 hierarchical downstream tasks ranging from preprocessing to specialized applications for each Sentinel mission.
- Copernicus-Embed-025deg: An embedding dataset that provides a global embedding map (721x1440x768) at 0.25°, integrating various sources of satellite observations at an extremely high compression ratio.
This dataset, model, and benchmark greatly improve the scalability, versatility, and multimodal adaptability of EO foundation models, while also creating new opportunities to connect EO, weather, and climate research.
Copernicus-Pretrain Dataset
Copernicus-Pretrain is an extension of the SSL4EO-S12 dataset to all major Sentinel missions (S1-S5P). The images are organized into ~310K regional grids (0.25°x0.25°, consistent with ERA5), densely covering the whole land surface and near-land ocean with time series from eight distinct Sentinel modalities.
Dataset Access
The dataset within this Hugging Face repository is available in two formats:
- Raw format (GeoTiff): This version is available directly via this Hugging Face dataset repository.
- Streaming format (WebDataset): This version is also available via this Hugging Face dataset repository.
For further details, refer to the Copernicus-Pretrain/ directory in the main GitHub repository.
Citation
If you use this dataset in your research, please cite the original paper:
@misc{wang2025unifiedcopernicusfoundationmodel,
title={Towards a Unified Copernicus Foundation Model for Earth Vision},
author={Yi Wang and Zhitong Xiong and Chenying Liu and Adam J. Stewart and Thomas Dujardin and Nikolaos Ioannis Bountos and Angelos Zavras and Franziska Gerken and Ioannis Papoutsis and Laura Leal-Taixé and Xiao Xiang Zhu},
year={2025},
eprint={2503.11849},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2503.11849},
}