LAE UAV dataset
收藏资源简介:
LAE UAV数据集由香港城市大学香港JC STEM智慧城市实验室与计算机科学系以及中国北京航空航天大学的研究团队创建,旨在模拟低空经济(LAE)中的真实城市GNSS拒止场景。该数据集包含357,690帧图像,其中包括对齐的RGB、深度和语义视图。数据集的创建过程采用了多摄像头无人机系统,用于捕获多方向视图,并提取紧凑的多视图特征。LAE UAV数据集适用于研究视觉导航、信息瓶颈、边缘推理和任务导向通信等领域,旨在解决低空经济中无人机在GPS信号不可用环境下的精确定位问题。
The LAE UAV Dataset was created by research teams from the Hong Kong JC STEM Smart City Laboratory and Department of Computer Science at City University of Hong Kong, as well as Beihang University in Beijing, China. It is intended to simulate realistic urban GNSS-denied scenarios in the Low-Altitude Economy (LAE). This dataset contains 357,690 image frames, including aligned RGB, depth and semantic views. The dataset was developed using a multi-camera UAV system to capture multi-directional views and extract compact multi-view features. The LAE UAV Dataset is applicable to research on visual navigation, information bottleneck, edge inference and task-oriented communication, and aims to address the precise positioning problem of UAVs in GPS-unavailable environments within the Low-Altitude Economy.




