Diverse Weather DroneVehicle
收藏资源简介:
Diverse Weather DroneVehicle是一个新增强的真实世界无人机基准数据集,由特文特大学和德雷塞尔大学的研究团队创建。该数据集通过标注天气条件将图像组织为四种场景:晴朗、黑暗、雾天和极端黑暗,旨在解决无人机目标检测在多样化天气条件下的领域泛化问题。数据来源于无人机拍摄的真实世界图像,经过人工标注天气标签以构建评估基准,为遥感社区提供了重要的跨天气场景领域泛化研究平台。
Diverse Weather DroneVehicle is a newly enhanced real-world drone benchmark dataset created by research teams from the University of Twente and Drexel University. The dataset organizes images into four scenarios according to annotated weather conditions: clear, dark, foggy, and extremely dark, with the goal of addressing the domain generalization issue of drone object detection under diverse weather conditions. Derived from real-world images captured by drones, the dataset features manually annotated weather labels to establish an evaluation benchmark, serving as an important research platform for the remote sensing community to carry out domain generalization studies across various weather scenarios.

- 1Bridge: Basis-Driven Causal Inference Marries VFMs for Domain Generalization特文特大学; 德雷塞尔大学 · 2026年



