蒸馏路网：跨传感器、分辨率和区域的通用路网提取
Distilled Roads: Generalisable Road Network Extraction Across Sensors, Resolutions, and Region
浏览论文内容
中文总结 AI 辅助
本研究提出结合跨分辨率知识蒸馏、多传感器训练与拓扑感知监督的框架，构建出泛化性强、效率高的路网提取模型，在公开基准上性能优于现有最优方法。
中文摘要 AI 辅助
从卫星影像中提取路网仍面临挑战，原因在于道路外观存在巨大地理差异、存在遮挡，以及不同分辨率和传感器引入的域偏移。现有模型通常在狭窄的分辨率-区域组合下训练，对未见过的环境（如农村场景、道路材质不同的区域或来自新卫星平台的影像）泛化能力较差，常产生断裂或不连通的预测结果。将这些模型适配到新域通常需要重新训练或微调，成本高昂且存在灾难性遗忘的风险。在本研究中，我们将全局道路提取重新定义为持续适配问题，而非架构问题。我们的框架结合了跨分辨率知识蒸馏（采用分辨率递减的课程学习）、多传感器训练以及拓扑感知监督，得到了一个可跨大洲多个卫星平台的0.3-1.0米影像进行泛化的单一模型。在公开基准（包括城市级和全球级）上，我们的模型相较于现有最优结果，F1值最高提升22个百分点，APLS值最高提升15个百分点，同时保持最高效，推理速度快3倍。我们的结果表明，通过针对性的训练策略（如数据课程、蒸馏和拓扑感知损失），而非日益复杂的架构，可实现对不同亚米级卫星影像的鲁棒性提升。
英文摘要
Road network segmentation from satellite imagery remains challenging due to large geographic variation in road appearance, occlusions, and domain shifts introduced by differing resolutions and sensors. Existing models, typically trained under narrow resolution--region combinations, generalise poorly to unseen environments such as rural settings, regions with distinct road materials, or imagery from new satellite platforms, often producing broken or disconnected predictions. Adapting these models to new domains usually requires retraining or fine-tuning, which is costly and risks catastrophic forgetting. In this work, we reframe global road extraction as a continual adaptation problem rather than an architectural one. Our framework combines cross-resolution knowledge distillation across a resolution-decreasing curriculum, multi-sensor training, and topology-aware supervision, yielding a single model that generalises across $0.3-1.0$ m imagery from multiple satellite platforms across continents. On publicly available benchmarks, including City-Scale and Global-Scale, our model outperforms state-of-the-art results by up to $22$ F1 points and $15$ APLS points, while remaining the most efficient, with $3\times$ faster inference. Our results suggest that improved robustness across diverse sub-meter satellite imagery can be achieved through targeted training strategies, such as data curricula, distillation, and topology-aware losses, rather than increasingly complex architectures.
