{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/ssl4eo-s12-a-large-scale-multi-modal-multi","title":"SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation","arxiv_id":"2211.07044","date":"2022-11-13","proceeding":null,"authors":["Yi Wang","Nassim Ait Ali Braham","Zhitong Xiong","Chenying Liu","Conrad M Albrecht","Xiao Xiang Zhu"],"abstract":"Self-supervised pre-training bears potential to generate expressive representations without human annotation. Most pre-training in Earth observation (EO) are based on ImageNet or medium-size, labeled remote sensing (RS) datasets. We share an unlabeled RS dataset SSL4EO-S12 (Self-Supervised Learning for Earth Observation - Sentinel-1/2) to assemble a large-scale, global, multimodal, and multi-seasonal corpus of satellite imagery from the ESA Sentinel-1 \\& -2 satellite missions. For EO applications we demonstrate SSL4EO-S12 to succeed in self-supervised pre-training for a set of methods: MoCo-v2, DINO, MAE, and data2vec. Resulting models yield downstream performance close to, or surpassing accuracy measures of supervised learning. In addition, pre-training on SSL4EO-S12 excels compared to existing datasets. We make openly available the dataset, related source code, and pre-trained models at https://github.com/zhu-xlab/SSL4EO-S12.","url_abs":"https://arxiv.org/abs/2211.07044v2","url_pdf":"https://arxiv.org/pdf/2211.07044v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"ssl4eo-s12-a-large-scale-multi-modal-multi","repo_url":"https://github.com/zhu-xlab/ssl4eo-s12","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"ssl4eo-s12-a-large-scale-multi-modal-multi","repo_url":"https://github.com/zhu-xlab/dino-mm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"ssl4eo-s12-a-large-scale-multi-modal-multi","repo_url":"https://github.com/zhu-xlab/softcon","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"ssl4eo-s12-a-large-scale-multi-modal-multi","repo_url":"https://github.com/zhu-xlab/ssl4eo-review","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"earth-observation","task_name":"Earth Observation"},{"task_slug":"multi-label-image-classification","task_name":"Multi-Label Image Classification"},{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"mae","method_name":"MAE"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[{"slug":"ssl4eo-s12","name":"SSL4EO-S12","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"MoCo-v2 (ResNet50, fine tune)","rank_in_archive_order":1,"of":10,"metrics":{"mAP (micro)":"91.8","official split":"No"},"uses_additional_data":true},{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"MoCo-v3 (ViT-S/16, fine tune)","rank_in_archive_order":2,"of":10,"metrics":{"mAP (micro)":"89.9","official split":"No"},"uses_additional_data":true},{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"MAE (ViT-S/16, fine tune)","rank_in_archive_order":4,"of":10,"metrics":{"mAP (micro)":"88.9","official split":"No"},"uses_additional_data":true},{"leaderboard":"/sota/multi-label-image-classification-on-2","task":"Multi-Label Image Classification","dataset":"BigEarthNet (official test set)","model":"MoCov3 (ViT-S/16)","rank_in_archive_order":2,"of":6,"metrics":{"F1 Score":"80.5","mAP (micro)":"89.3"},"uses_additional_data":true},{"leaderboard":"/sota/multi-label-image-classification-on-2","task":"Multi-Label Image Classification","dataset":"BigEarthNet (official test set)","model":"MoCov2 (ResNet50)","rank_in_archive_order":3,"of":6,"metrics":{"F1 Score":"79.8","mAP (micro)":"88.7"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2211.07044","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}