CARRADA
收藏资源简介:
CARRADA数据集是由法国国家科学研究中心创建,旨在为自动驾驶系统提供高质量的感知数据。该数据集包含同步的摄像头和雷达记录,以及范围-角度-多普勒注释,共有78个实例。数据集的创建过程采用了半自动注释方法,利用视觉信息进行雷达信号的标注。CARRADA数据集主要应用于自动驾驶领域,用于解决复杂驾驶场景中的对象检测、语义分割和跟踪问题。
The CARRADA dataset was developed by the French National Centre for Scientific Research (CNRS) to provide high-quality perception data for autonomous driving systems. It comprises synchronized camera and radar recordings alongside range-angle-Doppler annotations, with a total of 78 instances. A semi-automatic annotation approach was adopted during the dataset's construction, which leverages visual information to label radar signals. The CARRADA dataset is primarily applied in the autonomous driving field to address tasks such as object detection, semantic segmentation and tracking in complex driving scenarios.

- 1CARRADA Dataset: Camera and Automotive Radar with Range-Angle-Doppler Annotations法国国家科学研究中心 · 2021年



