发表机构
Microsoft AI for Good Research Lab（微软人工智能公益研究实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出Anaximander系统，统一模型来源与计算后端，通过交互界面和QGIS插件简化遥感深度学习应用，并演示了三个异构模型的无代码比较。
AI 中文摘要
在地理信息系统（GIS）内将深度学习模型应用于卫星图像，对遥感从业者而言仍存在高摩擦。模型以不兼容的格式出现，并针对不同的计算环境，从本地工作站到无服务器云服务。因此，每次评估都需要自定义部署、切片和地理配准代码，才能将首个预测结果送达分析人员的地图。这种摩擦阻碍了在模型选择直接影响运营成果（如地块划分、作物监测和灾害响应）的领域中进行系统性比较。我们提出了Anaximander，一个开源系统，将模型来源和计算位置选择统一在一个交互式界面之后。该系统的后端是一个推理服务器，可从多个常用来源加载模型，并在任何可访问的计算后端上提供服务。服务器提供会话管理和模型缓存，并逐块流式返回结果。后端与一个QGIS插件配对，该插件驱动切片、结果重组、地理配准和实时逐块状态可视化。另一条用户界面路径将图层图例作为提示注入视觉语言模型。我们在农业地块划分任务上，对三个异构模型进行了无代码并排比较，以演示该系统：通过云API的gpt-image-1、远程GPU上的Segment Anything Model 3（SAM3）以及本地CPU上的DelineateAnything。推理后端和协议是开源的，可在https://this https URL获取。
英文摘要
Applying deep learning models to satellite imagery from within geographic information systems (GIS) remains high-friction for remote sensing practitioners. Models arrive in incompatible formats and target different compute environments, from local workstations to serverless cloud services. As a result, every evaluation demands custom deployment, tiling, and georeferencing code before a single prediction reaches the analyst's map. This friction discourages systematic comparison in a domain where model choice directly affects operational outcomes such as field delineation, crop monitoring, and disaster response. We present Anaximander, an open-source system that unifies model source and compute location choice behind one interactive interface. The system's backend is an inference server that loads models from multiple commonly-used sources and serves them on any accessible compute backend. The server provides session management and model caching, and streams results back per tile. The backend is paired with a QGIS plugin that drives tiling, result reassembly, georeferencing, and real-time per-tile status visualization. An additional user-interface path injects layer legends as prompts into vision-language models. We demonstrate the system in a code-free side-by-side comparison of three heterogeneous models on an agricultural field delineation task: gpt-image-1 via a cloud API, Segment Anything Model 3 (SAM3) on a remote GPU, and DelineateAnything on a local CPU. The inference backend and protocol are open-source and available at https://github.com/microsoft/nxmndr.
CommentsAccepted as a poster at the TerraBytes II workshop, ECCV 2026
