CITR Dataset
收藏资源简介:
CITR数据集是在俄亥俄州立大学控制与智能交通研究实验室附近的一个停车场进行的受控实验中收集的。实验包括一系列基本的车群交互,旨在分析受车辆影响的交互式行人运动。使用带有下视摄像头的DJI Phantom 3 SE无人机作为记录设备。参与者是OSU CITR实验室的成员。在实验期间,他们被指示从一个小的区域(起点)走到另一个小的区域(目的地)。使用的车辆是由参与者手动驾驶的EZ-GO高尔夫球车。车辆顶部放置了3个标记,以帮助车辆运动跟踪,其中车辆中心位置通过几何计算得出。
The CITR dataset was collected during a controlled experiment conducted in a parking lot near the Control and Intelligent Transportation Research Laboratory at The Ohio State University. The experiment involved a series of basic vehicle-pedestrian interactions aimed at analyzing interactive pedestrian movements influenced by vehicles. A DJI Phantom 3 SE drone equipped with a downward-facing camera was used as the recording device. Participants were members of the OSU CITR laboratory. During the experiment, they were instructed to walk from a small area (starting point) to another small area (destination). The vehicle used was an EZ-GO golf cart manually driven by the participants. Three markers were placed on the top of the vehicle to assist in tracking its movement, with the vehicle's center position determined through geometric calculations.
数据集概述
数据集名称
- Vehicle-Crowd Intraction (VCI) - CITR Dataset
数据集内容
- Top-view trajectory data of pedestrians in groups under vehicle influence in controlled experiments
数据集更新
- Last updated on 04/19/2019
数据集采集地点
- Controlled experiments conducted in a parking lot near the Control and Intelligent Transportation Research (CITR) Lab at The Ohio State University (OSU)
数据集采集设备
- DJI Phantom 3 SE Drone with a down-facing camera on a gimbal system
数据集参与者
- Members of CITR Lab at OSU
实验车辆
- EZ-GO Golf Cart manually driven by a participant
实验配置
- Vehicle moves back and forth between two blue areas. Pedestrians move back and forth between two green areas. Interaction happens in the orange area.
数据集视频演示
数据集下载
- Stabilized raw videos can be downloaded from Google Drive and Baidu Yun (Code: h5kj)
数据集文件结构
datafolder contains rawrajectoriesandrajectories_filtered
数据集文件格式
- Filtered Trajectories:
.txt: Ratio/scale file for pixel to meter conversion.pdf: Overview plot of trajectories.csv: Pedestrian and vehicle trajectory data with headers
- Raw Trajectories:
.csv: Individual object trajectory data with specific headers
数据集统计信息
- 340 pedestrian trajectories
- 38 video clips, each about 20 seconds
数据集实验场景
- 6 different scenarios designed for detailed comparison and analysis
数据集使用建议
- Recommend using data in
rajectories_filtered
数据集联系方式
- E-mail: yang.3455@osu.edu
数据集引用
- If useful, consider citing the paper available at arXiv preprint




