用多模态知识扩展，让红外卫星实现全球降水精准反演。
From Swath to Full-Disc: Advancing Precipitation Retrieval with Multimodal Knowledge Expansion
- 分两阶段：先在扫描带内迁移多源数据知识，再拓展至全图域
- 在全图域上精度超越PERSIANN-CCS等主流产品，显著提升红外反演能力
- 适合遥感、气象预测领域研究者，尤其关注卫星降水反演的团队
近实时降水反演已因卫星技术得到提升。但基于红外的方法因与地表降水关联弱，精度较低；被动微波和雷达方法虽更准确，却受限于覆盖范围。为此提出降水反演扩展（PRE）任务，旨在实现红外卫星在扫描带外的全图域高精度降水反演。本文提出多模态知识扩展框架，包含两阶段流水线：在扫描带蒸馏阶段，通过协同掩码与小波增强（CoMWE），将多模态数据融合模型的知识迁移到红外模型中；在全图适应阶段，利用自掩码调优（Self-MaskTune）平衡多模态与全图红外知识，优化全域预测。在新构建的PRE基准测试上，PRE-Net显著优于PERSIANN-CCS、PDIR和IMERG等领先产品。代码将开源于https://github.com/Zjut-MultimediaPlus/PRE-Net。
原文摘要 · Abstract (English)
Accurate near-real-time precipitation retrieval has been enhanced by satellite-based technologies. However, infrared-based algorithms have low accuracy due to weak relations with surface precipitation, whereas passive microwave and radar-based methods are more accurate but limited in range. This challenge motivates the Precipitation Retrieval Expansion (PRE) task, which aims to enable accurate, infrared-based full-disc precipitation retrievals beyond the scanning swath. We introduce Multimodal Knowledge Expansion, a two-stage pipeline with the proposed PRE-Net model. In the Swath-Distilling stage, PRE-Net transfers knowledge from a multimodal data integration model to an infrared-based model within the scanning swath via Coordinated Masking and Wavelet Enhancement (CoMWE). In the Full-Disc Adaptation stage, Self-MaskTune refines predictions across the full disc by balancing multimodal and full-disc infrared knowledge. Experiments on the introduced PRE benchmark demonstrate that PRE-Net significantly advanced precipitation retrieval performance, outperforming leading products like PERSIANN-CCS, PDIR, and IMERG. The code will be available at https://github.com/Zjut-MultimediaPlus/PRE-Net.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。