FUSAR-GPT：面向 SAR 影像的时空特征嵌入与两阶段解耦视觉语言模型
FUSAR-GPT: A Spatiotemporal Feature-Embedded and Two-Stage Decoupled Visual Language Model for SAR Imagery
对做遥感/多模态的视觉研究者，可复用其「时空锚点」把多源时序特征注入视觉骨干网络以补偿 SAR 目标表征稀疏的思路，以及解耦知识注入与任务执行的两阶段 SFT 范式，可直接迁移到其他低质量或非 RGB 成像域的 VLM 适配。
原文依据 embeds multi-source remote-sensing temporal features into the model's visual backbone via 'spatiotemporal anchors'
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论对全天候、全时相合成孔径雷达（SAR）的智能解译研究，对推动遥感应用至关重要。近年来，尽管视觉语言模型（VLM）在 RGB 图像上展现出强大的开放世界理解能力，但由于成像机理复杂、对散射特征敏感以及高质量文本语料稀缺，将其直接应用于 SAR 领域时性能受到严重限制。为系统性地解决这一问题，我们构建了首个 SAR 图像-文本-AlphaEarth 特征三元组数据集，并研制了面向 SAR 的视觉语言模型 FUSAR-GPT。FUSAR-GPT 创新性地引入地理空间基线模型作为「世界知识」先验，并通过「时空锚点」将多源遥感时序特征嵌入模型的视觉骨干网络，从而对 SAR 图像中目标的稀疏表征实现动态补偿。此外，我们设计了两阶段 SFT 策略，将大模型的知识注入与任务执行解耦。时空特征嵌入与两阶段解耦范式使 FUSAR-GPT 在多项典型遥感视觉语言基准测试中取得最优性能，显著超越主流基线模型 10% 以上。
英文原文摘要 1354 字符
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 10%.