用低分辨率图+智能采样，高效识别卫星图像中的地物尺度。
Scale-Aware Recognition in Satellite Images under Resource Constraints
- 从高分辨率模型蒸馏知识到低分辨率模型，降低计算成本。
- 根据模型分歧动态采样高分辨率图像，减少76.3%使用量。
- 结合大模型推理概念尺度，适合资源受限的遥感应用。
卫星图像中地物（如森林、泳池）的识别高度依赖空间尺度和图像分辨率，带来两大挑战：如何确定最佳识别分辨率，以及何时何地应获取更昂贵的高分辨率（HR）图像。本文提出一种新方案，包含三部分：(1) 将在高分辨率图像上训练的模型知识蒸馏至低分辨率（LR）模型；(2) 基于模型分歧的高分辨率图像采样策略；(3) 利用大语言模型（LLM）推断概念的“尺度”信息。该系统实现高效的尺度感知识别，在预算约束下相比单尺度推理提升性能，较全高分辨率基线最高提升26.3%，同时仅使用76.3%的高分辨率图像。
原文摘要 · Abstract (English)
Recognition of features in satellite imagery (forests, swimming pools, etc.) depends strongly on the spatial scale of the concept and therefore the resolution of the images. This poses two challenges: Which resolution is best suited for recognizing a given concept, and where and when should the costlier higher-resolution (HR) imagery be acquired? We present a novel scheme to address these challenges by introducing three components: (1) A technique to distill knowledge from models trained on HR imagery to recognition models that operate on imagery of lower resolution (LR), (2) a sampling strategy for HR imagery based on model disagreement, and (3) an LLM-based approach for inferring concept "scale". With these components we present a system to efficiently perform scale-aware recognition in satellite imagery, improving accuracy over single-scale inference while following budget constraints. Our novel approach offers up to a 26.3% improvement over entirely HR baselines, using 76.3% fewer HR images.
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。