Video presentation

We present a novel method for road intersection detection and localization using on-vehicle LiDAR.

Our approach leverages semantic information and vehicle odometry as inputs to detect intersection points in a bird’s-eye-view (BEV) representation. For evaluation, we introduce an automated pipeline that pairs localized intersection points with OpenStreetMap (OSM) intersection nodes using precise GNSS/INS ground-truth poses.

Experiments on the SemanticKITTI dataset show that our method outperforms the latest learning-based baselines in accuracy and reliability. Sensitivity tests demonstrate the method’s robustness to challenging segmentation errors, highlighting its real-world applicability.

Publications

  • Nguyen Hoang Khoi Tran, Julie Stephany Berrio, Mao Shan, Zhenxing Ming, and Stewart Worrall, “InterLoc: Lidar-based intersection localization using road segmentation with automated evaluation method,” in IEEE International Conference on Intelligent Transportation Systems (ITSC), Gold Coast, Australia, 2025. Available here, [doi]

Citing

If you find this work useful please cite

@inproceedings{tran2025interloc,
  author    = {Tran, Nguyen Hoang Khoi and Berrio, Julie Stephany and Shan, Mao and Ming, Zhenxing and Worrall, Stewart},
  title     = {{InterLoc}: {LiDAR}-based Intersection Localization using Road Segmentation with Automated Evaluation Method},
  booktitle = {2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC)},
  year      = {2025},
  address   = {Gold Coast, Australia}
}

Updated: