Intersection Vehicle Trajectory Prediction Data

Citation Author(s):
yue
liu
Submitted by:
yue liu
Last updated:
Sat, 12/02/2023 - 21:50
DOI:
10.21227/67hf-9h06
License:
229 Views
Categories:
Keywords:
0
0 ratings - Please login to submit your rating.

Abstract 

We have selected three intersections within Xi'an, Shaanxi Province, China for data collection. These include the intersection of Taibai South Road and Keji Road, the intersection of Chang'an North Road and Yanta West Road, and the intersection of Yanta North Road and Youyi Road. The UAV for the purpose of video data collection is employed to investigate intersections, with a high-definition video resolution of 4096 * 2160. The spatiotemporal position information of target vehicles was tracked and labeled using Datafromsky software. The video images are geographically calibrated in the Universal Transverse Mercator grid system (UTM) using the four-point calibration method.

Instructions: 

The video data is extracted from the software DataFromSky in order to obtain vehicle trajectories, and it is saved in CSV format. Due to the maximum duration of 5 minutes for each video captured by the unmanned aerial vehicle (UAV), the data is segmented into multiple segments. Subsequently, these segments are organized and classified into a comprehensive dataset, including categories for straight movements as well as right and left turns.

Comments

Dear Sir,

 

I am Dr. Jamal from Qassim University, Saudi Arabia. I am currently working on a similar project related to vehicle trajectory prediction. I would appreciate it if you can please share the above (ITVP.zip) data (email: ajjamal@qec.edu.sa) as well the source code for reconstructing or projecting vehicle trajectories.  I will provide proper attribution in case we use for our ongoing or future works. I look forward to your feedback. 

 

Best Regards,

Jamal

Submitted by Arshad Jamal on Mon, 03/25/2024 - 05:10