AirV2X-Perception

该数据集是为了支持无人机辅助的车辆与一切事物（V2X）协同感知研究而设计的大规模数据集，包含了在城市、郊区和乡村环境中，跨越6.73小时的驾驶场景，这些场景涵盖了不同的天气和光照条件。数据集支持三种无人机导航策略：悬停、巡逻和护航，并且传感器数据的采集频率为5赫兹。该数据集包含了1,961,484个标注对象，涵盖多个类别，旨在用于开发和评估车辆到无人机（V2D）算法。

This large-scale dataset is designed to support research on unmanned aerial vehicle (UAV)-assisted vehicle-to-everything (V2X) collaborative perception. It includes driving scenarios across urban, suburban and rural environments, with a total duration of 6.73 hours, and these scenarios cover a wide range of weather and lighting conditions. The dataset supports three UAV navigation strategies: hovering, patrolling and escorting, with sensor data collected at a frequency of 5 Hz. It contains 1,961,484 annotated objects across multiple categories, and is intended for the development and evaluation of vehicle-to-drone (V2D) algorithms.

搜集汇总

背景与挑战

背景概述

AirV2X-Perception是一个专注于统一空中-地面/车辆到万物（V2X）协作感知的数据集，可从Hugging Face获取，并支持多种模型训练和评估配置。