THOR模型统一处理多源遥感数据，支持灵活分辨率推理。
THOR: A Versatile Foundation Model for Earth Observation Climate and Society Applications
- 设计可适应计算资源的架构，统一处理10~1000米分辨率的多卫星数据
- 在10%数据量下仍达领先性能，验证其在小样本场景优势
- 适合气候与社会应用中需动态调整算力与精度的研究者
当前地球观测基础模型架构僵化，难以应对异构传感器和固定块大小限制，制约实际部署中计算与精度的灵活权衡。本文提出THOR，一种“计算自适应”基础模型，同时解决输入异质性与部署刚性问题。THOR是首个统一处理欧空局哥白尼计划哨兵-1、-2、-3（OLCI与SLSTR）卫星数据的架构，可直接处理其原生10米至1000米分辨率数据。通过创新的随机块大小与输入图像尺寸预训练策略，单组预训练权重即可在推理阶段任意调整块大小，实现无需重训的计算成本与特征分辨率动态平衡。我们在新构建的大规模多传感器数据集THOR Pretrain上预训练THOR，下游基准测试表现达到当前最优，尤其在数据受限场景如PANGAEA 10%划分下表现突出，验证了其灵活特征生成对气候与社会应用的强大适用性。
原文摘要 · Abstract (English)
Current Earth observation foundation models are architecturally rigid, struggle with heterogeneous sensors and are constrained to fixed patch sizes. This limits their deployment in real-world scenarios requiring flexible computeaccuracy trade-offs. We propose THOR, a "computeadaptive" foundation model that solves both input heterogeneity and deployment rigidity. THOR is the first architecture to unify data from Copernicus Sentinel-1, -2, and -3 (OLCI & SLSTR) satellites, processing their native 10 m to 1000 m resolutions in a single model. We pre-train THOR with a novel randomized patch and input image size strategy. This allows a single set of pre-trained weights to be deployed at inference with any patch size, enabling a dynamic trade-off between computational cost and feature resolution without retraining. We pre-train THOR on THOR Pretrain, a new, large-scale multi-sensor dataset and demonstrate state-of-the-art performance on downstream benchmarks, particularly in data-limited regimes like the PANGAEA 10% split, validating that THOR's flexible feature generation excels for diverse climate and society applications.
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。