FUSAR-GPT : A Spatiotemporal Feature-Embedded and Two-Stage Decoupled Visual Language Model for SAR Imagery
This addresses the need for intelligent interpretation of SAR imagery in remote sensing applications, representing a domain-specific advancement.
The paper tackled the problem of applying Visual Language Models (VLMs) to Synthetic Aperture Radar (SAR) imagery, which suffers from imaging complexity and data scarcity, by developing FUSAR-GPT with spatiotemporal feature embedding and a two-stage decoupling strategy, achieving state-of-the-art performance with over 12% improvement on remote sensing benchmarks.
Research on the intelligent interpretation of all-weather, all-time Synthetic Aperture Radar (SAR) is crucial for advancing remote sensing applications. In recent years, although Visual Language Models (VLMs) have demonstrated strong open-world understanding capabilities on RGB images, their performance is severely limited when directly applied to the SAR field due to the complexity of the imaging mechanism, sensitivity to scattering features, and the scarcity of high-quality text corpora. To systematically address this issue, we constructed the inaugural SAR Image-Text-AlphaEarth feature triplet dataset and developed FUSAR-GPT, a VLM specifically for SAR. FUSAR-GPT innovatively introduces a geospatial baseline model as a 'world knowledge' prior and embeds multi-source remote-sensing temporal features into the model's visual backbone via 'spatiotemporal anchors', enabling dynamic compensation for the sparse representation of targets in SAR images. Furthermore, we designed a two-stage SFT strategy to decouple the knowledge injection and task execution of large models. The spatiotemporal feature embedding and the two-stage decoupling paradigm enable FUSAR-GPT to achieve state-of-the-art performance across several typical remote sensing visual-language benchmark tests, significantly outperforming mainstream baseline models by over 12%.