Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Automated Construction of Time-Space Diagrams for Traffic Analysis Using Street-View Video Sequence

Tanay Rastogi, Mårten Björkman

TL;DR

This work addresses the challenge of generating time-space diagrams with maximal spatial and temporal coverage by using street-view video sequences captured from a moving vehicle. It combines YOLOv5-based detection, StrongSORT tracking, and a photogrammetry/geodesy distance-estimation pipeline to infer vehicle trajectories and produce time-space diagrams, evaluated on the KITTI dataset. Results show the approach can approximate ground-truth trajectories and traffic patterns, with errors primarily stemming from detector/tracker performance and distance estimation; these can be mitigated by model re-training and integrating additional data sources like LiDAR. The proposed method offers a flexible, scalable avenue for vehicle trajectory analysis and traffic-state estimation, with potential benefits for infrastructure design, traffic management, and broader urban mobility research.

Abstract

Time-space diagrams are essential tools for analyzing traffic patterns and optimizing transportation infrastructure and traffic management strategies. Traditional data collection methods for these diagrams have limitations in terms of temporal and spatial coverage. Recent advancements in camera technology have overcome these limitations and provided extensive urban data. In this study, we propose an innovative approach to constructing time-space diagrams by utilizing street-view video sequences captured by cameras mounted on moving vehicles. Using the state-of-the-art YOLOv5, StrongSORT, and photogrammetry techniques for distance calculation, we can infer vehicle trajectories from the video data and generate time-space diagrams. To evaluate the effectiveness of our proposed method, we utilized datasets from the KITTI computer vision benchmark suite. The evaluation results demonstrate that our approach can generate trajectories from video data, although there are some errors that can be mitigated by improving the performance of the detector, tracker, and distance calculation components. In conclusion, the utilization of street-view video sequences captured by cameras mounted on moving vehicles, combined with state-of-the-art computer vision techniques, has immense potential for constructing comprehensive time-space diagrams. These diagrams offer valuable insights into traffic patterns and contribute to the design of transportation infrastructure and traffic management strategies.

Automated Construction of Time-Space Diagrams for Traffic Analysis Using Street-View Video Sequence

TL;DR

This work addresses the challenge of generating time-space diagrams with maximal spatial and temporal coverage by using street-view video sequences captured from a moving vehicle. It combines YOLOv5-based detection, StrongSORT tracking, and a photogrammetry/geodesy distance-estimation pipeline to infer vehicle trajectories and produce time-space diagrams, evaluated on the KITTI dataset. Results show the approach can approximate ground-truth trajectories and traffic patterns, with errors primarily stemming from detector/tracker performance and distance estimation; these can be mitigated by model re-training and integrating additional data sources like LiDAR. The proposed method offers a flexible, scalable avenue for vehicle trajectory analysis and traffic-state estimation, with potential benefits for infrastructure design, traffic management, and broader urban mobility research.

Abstract

Time-space diagrams are essential tools for analyzing traffic patterns and optimizing transportation infrastructure and traffic management strategies. Traditional data collection methods for these diagrams have limitations in terms of temporal and spatial coverage. Recent advancements in camera technology have overcome these limitations and provided extensive urban data. In this study, we propose an innovative approach to constructing time-space diagrams by utilizing street-view video sequences captured by cameras mounted on moving vehicles. Using the state-of-the-art YOLOv5, StrongSORT, and photogrammetry techniques for distance calculation, we can infer vehicle trajectories from the video data and generate time-space diagrams. To evaluate the effectiveness of our proposed method, we utilized datasets from the KITTI computer vision benchmark suite. The evaluation results demonstrate that our approach can generate trajectories from video data, although there are some errors that can be mitigated by improving the performance of the detector, tracker, and distance calculation components. In conclusion, the utilization of street-view video sequences captured by cameras mounted on moving vehicles, combined with state-of-the-art computer vision techniques, has immense potential for constructing comprehensive time-space diagrams. These diagrams offer valuable insights into traffic patterns and contribute to the design of transportation infrastructure and traffic management strategies.
Paper Structure (12 sections, 4 equations, 4 figures, 5 tables)

This paper contains 12 sections, 4 equations, 4 figures, 5 tables.

Table of Contents

  1. INTRODUCTION
  2. Methodology
  3. Dataset
  4. Object Detection
  5. Multi-Object Tracking
  6. Lane Distance Estimation
  7. Results
  8. Object Detection Results
  9. Multi-Object Tracking Results
  10. Lane Distance Estimation Results
  11. Time-space Diagram Results
  12. Conclusion

Figures (4)

  • Figure 1: Flow chart that illustrates the proposed methodology of extracting the vehicle trajectory from a street-view video sequence using different computer vision algorithms.
  • Figure 2: KITTI image frame showcasing the bounding boxes (in RED) and tracking IDs (in GREEN) for vehicles in the opposite lane derived using YOLOv5m and StrongSORT.
  • Figure 3: RMSE distribution between ground truth depth and camera distance in two scenarios: RMSE (Pred) (in ORANGE) uses YOLOv5m bounding boxes for calculation, while RMSE (Calc, GT) (in BLUE) uses ground truth bounding boxes.
  • Figure 4: Time-space diagram generated for video 0004.mp4 from KITTI dataset with true trajectories (in RED) and predicted trajectories (in BLUE).