统一地球嵌入数据标准，让遥感模型更易复现与比较
Earth Embeddings as Products: Taxonomy, Ecosystem, and Standardized Access
- 提出数据-工具-价值三层分类框架，梳理现有遥感嵌入产品
- 发现不同嵌入数据格式不兼容，导致模型对比困难
- 扩展TorchGeo实现统一加载接口，支持多种嵌入产品
地理空间基础模型（GFMs）提供强大表征，但高计算成本限制其广泛应用。预计算的嵌入数据产品作为实用的“冻结”替代方案，却存在于格式和分辨率不兼容的碎片化生态中。这种缺乏标准化的现象造成工程瓶颈，阻碍了模型的有效比较与可复现性。本文通过三层次分类体系——数据、工具、价值，系统梳理现有嵌入产品，并识别互操作性障碍。为弥合此差距，我们扩展TorchGeo，构建统一API以标准化多样嵌入产品的加载与查询。将嵌入视为第一类地理空间数据集，实现下游分析与模型特定工程的解耦，为更透明、可访问的地球观测工作流提供路线图。
原文摘要 · Abstract (English)
Geospatial Foundation Models (GFMs) provide powerful representations, but high compute costs hinder their widespread use. Pre-computed embedding data products offer a practical "frozen" alternative, yet they currently exist in a fragmented ecosystem of incompatible formats and resolutions. This lack of standardization creates an engineering bottleneck that prevents meaningful model comparison and reproducibility. We formalize this landscape through a three-layer taxonomy: Data, Tools, and Value. We survey existing products to identify interoperability barriers. To bridge this gap, we extend TorchGeo with a unified API that standardizes the loading and querying of diverse embedding products. By treating embeddings as first-class geospatial datasets, we decouple downstream analysis from model-specific engineering, providing a roadmap for more transparent and accessible Earth observation workflows.
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。