OceanSAR-2: A Universal Feature Extractor for SAR Ocean Observation

TL;DR

OceanSAR-2 is a new foundation model for SAR ocean observation that improves performance and reduces training costs through advanced self-supervised learning and data curation, enabling versatile ocean monitoring applications.

Contribution

It introduces OceanSAR-2, a universal SAR feature extractor with enhanced training methods and benchmark datasets for ocean observation tasks.

Findings

Strong transfer performance across multiple ocean tasks

Reduced training costs compared to previous models

Provides standardized datasets for SAR ocean research

Abstract

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.