JRDB-Traj
收藏资源简介:
JRDB-Traj数据集是由洛桑联邦理工学院和莫纳什大学合作创建的,旨在为人群中的轨迹预测提供一个全面的基准。该数据集包含54个序列,涵盖室内外场景,从机器人的视角提供了包括所有代理位置、场景图像和点云在内的详细数据。数据集的创建过程涉及使用3D边界框注释和跟踪ID,以生成轨迹。JRDB-Traj数据集的应用领域主要集中在自主导航系统,特别是机器人和自动驾驶车辆,旨在通过预测未来位置来提高导航的安全性和效率。
JRDB-Traj dataset was co-developed by École Polytechnique Fédérale de Lausanne (EPFL) and Monash University, aiming to serve as a comprehensive benchmark for trajectory prediction in crowded scenarios. The dataset contains 54 sequences covering both indoor and outdoor scenes, and provides detailed data including positions of all agents, scene images and point clouds from the robot's perspective. The creation process of the dataset utilizes 3D bounding box annotations and tracking IDs to generate trajectories. The application scenarios of JRDB-Traj dataset mainly focus on autonomous navigation systems, particularly robots and autonomous vehicles, with the goal of enhancing the safety and efficiency of navigation by predicting future positions of agents.

- 1JRDB-Traj: A Dataset and Benchmark for Trajectory Forecasting in Crowds洛桑联邦理工学院, 莫纳什大学 · 2023年



