Saved in:
Bibliographic Details
Main Authors: Tuel, Alexandre, Kerdreux, Thomas, Febvre, Quentin, Mouche, Alexis, Grouazel, Antoine, Miadana, Jean-Renaud, Audras, Antoine, Wang, Chen, Chapron, Bertrand
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.07392
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912817970413568
author Tuel, Alexandre
Kerdreux, Thomas
Febvre, Quentin
Mouche, Alexis
Grouazel, Antoine
Miadana, Jean-Renaud
Audras, Antoine
Wang, Chen
Chapron, Bertrand
author_facet Tuel, Alexandre
Kerdreux, Thomas
Febvre, Quentin
Mouche, Alexis
Grouazel, Antoine
Miadana, Jean-Renaud
Audras, Antoine
Wang, Chen
Chapron, Bertrand
contents We present OceanSAR-2, the second generation of our foundation model for SAR-based ocean observation. Building on our earlier release, which pioneered self-supervised learning on Sentinel-1 Wave Mode data, OceanSAR-2 relies on improved SSL training and dynamic data curation strategies, which enhances performance while reducing training cost. OceanSAR-2 demonstrates strong transfer performance across downstream tasks, including geophysical pattern classification, ocean surface wind vector and significant wave height estimation, and iceberg detection. We release standardized benchmark datasets, providing a foundation for systematic evaluation and advancement of SAR models for ocean applications.
format Preprint
id arxiv_https___arxiv_org_abs_2601_07392
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle OceanSAR-2: A Universal Feature Extractor for SAR Ocean Observation
Tuel, Alexandre
Kerdreux, Thomas
Febvre, Quentin
Mouche, Alexis
Grouazel, Antoine
Miadana, Jean-Renaud
Audras, Antoine
Wang, Chen
Chapron, Bertrand
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
We present OceanSAR-2, the second generation of our foundation model for SAR-based ocean observation. Building on our earlier release, which pioneered self-supervised learning on Sentinel-1 Wave Mode data, OceanSAR-2 relies on improved SSL training and dynamic data curation strategies, which enhances performance while reducing training cost. OceanSAR-2 demonstrates strong transfer performance across downstream tasks, including geophysical pattern classification, ocean surface wind vector and significant wave height estimation, and iceberg detection. We release standardized benchmark datasets, providing a foundation for systematic evaluation and advancement of SAR models for ocean applications.
title OceanSAR-2: A Universal Feature Extractor for SAR Ocean Observation
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.07392