ADUULM-360
收藏资源简介:
ADUULM-360数据集是由乌尔姆大学测量、控制和微技术研究所创建的多模态深度估计数据集,旨在解决自动驾驶中的深度感知问题。该数据集包含约100万张摄像头图像、25万次激光雷达点云扫描和80万次雷达点云扫描,覆盖了从良好到恶劣的多种天气条件。数据集的创建过程包括传感器校准、数据处理和隐私保护措施,确保数据的高质量和可用性。该数据集主要应用于自动驾驶领域,特别是深度估计和环境感知,旨在提高自动驾驶系统在不同天气条件下的鲁棒性和准确性。
The ADUULM-360 dataset is a multimodal depth estimation dataset developed by the Institute of Measurement, Control and Microtechnology at Ulm University, aiming to address depth perception challenges in autonomous driving. This dataset encompasses approximately 1 million camera images, 250,000 LiDAR point cloud scans and 800,000 radar point cloud scans, covering a wide range of weather conditions from favorable to harsh. The dataset's development process includes sensor calibration, data processing and privacy protection measures, ensuring the high quality and availability of the collected data. It is primarily applied in the autonomous driving domain, particularly for depth estimation and environmental perception, with the goal of improving the robustness and accuracy of autonomous driving systems under various weather conditions.
ADUULM-360 数据集
概述
- 名称: ADUULM-360 数据集
- 类型: 多模态数据集
- 应用: 深度估计
- 环境: 恶劣天气
发布信息
- 论文: ITSC 2024
- 状态: 数据集访问和基线模型源代码即将发布

- 1The ADUULM-360 Dataset -- A Multi-Modal Dataset for Depth Estimation in Adverse Weather乌尔姆大学测量、控制和微技术研究所 · 2024年



