aiMotive Dataset

aiMotive Dataset是一个专为自动驾驶设计的多元数据集，包含176个场景，这些场景通过同步和校准的激光雷达（LiDAR）、摄像头和雷达传感器捕捉，覆盖360度视野。该数据集在高速公路、城市和郊区等多样化地理区域以及白天、夜晚和雨天等不同时间和天气条件下收集。数据集提供3D边界框，具有跨帧一致的标识符，适用于3D物体检测和多物体跟踪及预测任务。此外，数据集还支持多模态传感器融合和稳健的长距离感知系统的研究，旨在解决自动驾驶中的安全性和感知范围问题。

The aiMotive Dataset is a diverse dataset specifically designed for autonomous driving. It contains 176 scenarios captured by synchronized and calibrated LiDAR, camera, and radar sensors, with a 360-degree field of view. The dataset was collected across diverse geographic regions including highways, urban areas, and suburbs, as well as under varying time and weather conditions such as daytime, nighttime, and rainy days. It provides 3D bounding boxes with consistent cross-frame identifiers, which are applicable to 3D object detection, multi-object tracking and prediction tasks. Additionally, the dataset supports research on multimodal sensor fusion and robust long-range perception systems, aiming to address the safety and perception range issues in autonomous driving.