发表机构
Fraunhofer Institute IVI; Technical University of Munich; Munich Center for Machine Learning (MCML); University of Cambridge; Univ. of Applied Sciences Ingolstadt（弗劳恩霍夫交通系统研究所; 慕尼黑工业大学; 慕尼黑机器学习中心; 剑桥大学; 英戈尔施塔特应用科学大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
MultiFly是一个真实世界低空多模态航空数据集，通过仅标注115张RGB图像并利用共享几何表示将标签高效迁移至RGB、热成像、LiDAR和雷达模态，实现高一致性的语义分割基准，为多模态航空感知提供可扩展基础。
AI 中文摘要
我们介绍了MultiFly，一个用于RGB、热成像、LiDAR和雷达模态语义感知的真实世界低空无人机（UAV）数据集。MultiFly提供了来自四个郊区场景的17,272个同步样本，包含15个语义类别的逐帧标注，以及标定和GNSS-RTK/IMU测量数据。为了避免昂贵且不一致的模态特定标注，我们仅通过共享的几何表示将115张手动标注的RGB图像的标签传播到所有四种模态。该方法为额外的17,157张RGB图像、17,272张热成像图像、8.4亿个LiDAR点和340万个雷达点生成了语义标签。迁移后的标注与保留的手动标注平均一致性达到89.93%，所有六对模态间的平均语义一致性达到90.94%。我们进一步为所有四种模态建立了语义分割基准，揭示了密集LiDAR和稀疏雷达数据的不同架构行为。总的来说，MultiFly为多模态航空感知提供了一个可扩展的基础，并且据我们所知，是第一个将RGB、热成像、LiDAR和雷达的一致逐帧语义标注相结合的公开真实世界低空航空基准。数据可在以下网址获取：https URL。
英文摘要
We introduce MultiFly, a real-world, low-altitude UAV dataset for semantic perception across RGB, thermal, LiDAR, and radar modalities. MultiFly provides 17,272 synchronized samples from four suburban scenes with frame-wise annotations for 15 semantic classes, together with calibration and GNSS-RTK/IMU measurements. To avoid costly and inconsistent modality-specific annotation, we propagate labels from only 115 manually annotated RGB images through shared geometric representations to all four modalities. This approach generates semantic labels for 17,157 additional RGB images, 17,272 thermal images, 840M LiDAR points, and 3.4M radar points. Transferred annotations achieve 89.93% average agreement with held-out manual annotations, and 90.94% average semantic consistency across all six modality pairs. We further establish semantic segmentation benchmarks for all four modalities, revealing distinct architectural behavior for dense LiDAR and sparse radar data. Taken together, MultiFly provides a scalable foundation for multimodal aerial perception and, to the best of our knowledge, the first public real-world low-altitude aerial benchmark that combines consistent frame-wise semantic annotations for RGB, thermal, LiDAR, and radar. Data at https://github.com/markus-42/multifly.
