RadarScenes 雷达点云数据集

RadarScenes 是关于汽车应用的真实世界雷达点云数据集，由 4 个小时的驾驶测量结果和逐点标注组成。数据由测试车上的 4 个系列的雷达传感器收集，7,500 多个动态对象的单独检测结果被人工分组并进行标注。该数据集的目的是开发基于机器学习的新型雷达感知算法，重点关注移动道路用户。

RadarScenes is a real-world radar point cloud dataset for automotive applications, consisting of 4 hours of driving measurements and per-point annotations. The data was collected by four series of radar sensors mounted on a test vehicle, and over 7,500 individual detections of dynamic objects were manually grouped and annotated. The purpose of this dataset is to develop novel machine learning-based radar perception algorithms, with a focus on moving road users.

搜集汇总

背景与挑战

背景概述

RadarScenes是一个用于汽车应用的真实世界雷达点云数据集，包含4小时的驾驶测量数据和7,500多个动态对象的逐点标注，旨在支持移动道路用户的雷达感知算法开发。数据集由4个雷达传感器收集，适用于非商业用途。