SCOPE
收藏资源简介:
SCOPE数据集是由图宾根大学等机构创建的合成多模态数据集,专门用于集体感知研究。该数据集包含17,600帧,涵盖超过40种不同场景,包括复杂的天气条件和多种道路使用者。数据集通过使用真实的相机和LiDAR模型以及物理上准确的天气模拟来增强其真实性。创建过程中,数据集结合了CARLA模拟器和改进的LiDAR传感器模型,以确保数据的多样性和真实性。SCOPE数据集主要应用于自动驾驶领域,旨在解决感知范围受限和环境条件影响的问题,提高自动驾驶车辆的安全性和性能。
The SCOPE dataset is a synthetic multimodal dataset created by the University of Tübingen and other institutions, specifically dedicated to collective perception research. This dataset contains 17,600 frames, covering more than 40 distinct scenarios including complex weather conditions and various road users. It enhances its realism by utilizing realistic camera and LiDAR models as well as physically accurate weather simulations. During its development, the dataset integrates the CARLA simulator and an improved LiDAR sensor model to ensure data diversity and realism. The SCOPE dataset is primarily applied in the autonomous driving domain, aiming to address issues such as limited perception range and the impact of environmental conditions, thereby improving the safety and performance of autonomous vehicles.

- 1SCOPE: A Synthetic Multi-Modal Dataset for Collective Perception Including Physical-Correct Weather Conditions图宾根大学,科学学院,计算机科学系,嵌入式系统小组 · 2024年



