CLOSP: A Unified Semantic Space for SAR, MSI, and Text in Remote Sensing

TL;DR

This paper introduces CLOSP, a unified semantic embedding space for SAR, multispectral, and text data in remote sensing, enabling improved retrieval and understanding of satellite imagery across sensors and applications.

Contribution

The paper presents CLOSP, a novel framework that aligns SAR, multispectral, and text data into a shared semantic space, advancing cross-modal retrieval and interpretation in remote sensing.

Findings

CLOSP achieves 54% improvement in retrieval performance over existing models.

Unified training transfers semantic knowledge from optical to SAR imagery.

GeoCLOSP effectively incorporates geographic context for location-specific tasks.

Abstract

Retrieving relevant imagery from vast satellite archives is crucial for applications like disaster response and long-term climate monitoring. However, most text-to-image retrieval systems are limited to RGB data, failing to exploit the unique physical information captured by other sensors, such as the all-weather structural sensitivity of Synthetic Aperture Radar (SAR) or the spectral signatures in optical multispectral data. To bridge this gap, we introduce CrisisLandMark, a new large-scale corpus of over 647,000 Sentinel-1 SAR and Sentinel-2 multispectral images paired with structured textual annotations for land cover, land use, and crisis events harmonized from authoritative land cover systems (CORINE and Dynamic World) and crisis-specific sources. We then present CLOSP (Contrastive Language Optical SAR Pretraining), a novel framework that uses text as a bridge to align unpaired optical and SAR images into a unified embedding space. Our experiments show that CLOSP achieves a new state-of-the-art, improving retrieval nDGC@1000 by 54% over existing models. Additionally, we find that the unified training strategy overcomes the inherent difficulty of interpreting SAR imagery by transferring rich semantic knowledge from the optical domain with indirect interaction. Furthermore, GeoCLOSP, which integrates geographic coordinates into our framework, creates a powerful trade-off between generality and specificity: while the CLOSP excels at general semantic tasks, the GeoCLOSP becomes a specialized expert for retrieving location-dependent crisis events and rare geographic features. This work highlights that the integration of diverse sensor data and geographic context is essential for unlocking the full potential of remote sensing archives.