Monitoring-Based Traffic Participant Detection in Urban Mixed Traffic: A Novel Dataset and A Tailored Detector

Monitoring-based traffic participant detection (TPD) is a highly desirable but challenging task. So far, deep learning-based methods have attained significant improvements on the TPD task, but oftentimes fail in urban mixed traffic due to the lack of relevant datasets and suitable detectors. In this study, the authors propose a large and detailed dataset named SEU_PML specialized for monitoring-based TPD in urban mixed traffic. This dataset contains a total of 270,684 objects annotated with 2D bounding box and covers 13 sub-categories, having (i) high-resolution images (from1920 x 1080 to 4096 x 2160 pixels), (ii) high-quality annotation (annotation accuracy reaches 98%), and (iii) rich traffic scenarios covering diverse traffic scenes as well as different weather and illumination conditions. The mixed traffic along with high-quality annotation bring about a variety of small objects. To further address the issue on small object detection, they propose a novel detector named YOLO SOD, which embeds a super-resolution feature extraction module and uses knowledge distillation to learn the knowledge how the detector with high-resolution inputs perceives small objects. Moreover, a novel loss function named S-IoU is designed to enable YOLO SOD to focus more on small objects. Experimental results show that (1) the YOLO SOD detector has an increased mAP of 1.58% and operates approximately four times faster when compared to a state-of-art detector; (2) the detectors trained on the SEU_PML dataset have a strong transferability and could be well applied to traffic participant detection in urban mixed traffic. The authors' dataset is now available at https://github.com/vvgoder/SEU_PML_Dataset.

• Record URL:
  • https://doi.org/10.1109/TITS.2023.3304288
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/41297384
• Supplemental Notes:
  • Copyright © 2024, IEEE.
• Authors:
  • Zhou, Wei
  • Wang, Chen
  • Xia, Jingxin
  • Qian, Zhendong
  • Wu, Yuan
• Publication Date: 2024-1

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: pp 189-202
• Serial:
  • IEEE Transactions on Intelligent Transportation Systems
  • Volume: 25
  • Issue Number: 1
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 1524-9050
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6979

Subject/Index Terms

• TRT Terms: Datasets; Lighting; Urban areas; Vehicle detectors; Vehicle mix; Weather
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 01918202
• Record Type: Publication
• Files: TRIS
• Created Date: May 10 2024 4:51PM