Video presentation

We propose a novel framework that leverages road intersections as distinctive landmarks for vehicle global localization on OpenStreetMap.

Our method constructs compact binary descriptors by jointly encoding road and building imprints from point clouds and OpenStreetMap priors. To bridge modality gaps, we introduce discrepancy mitigation, orientation determination, and area-equalized sampling strategies, enabling robust cross-modal matching.

Experiments on the KITTI dataset demonstrate that our method achieves state-of-the-art accuracy, outperforming recent baselines by a large margin. The framework generalizes to sensors that can produce dense structural point clouds, offering a scalable and cost-effective solution for robust vehicle localization.

Publications

  • Nguyen Hoang Khoi Tran, Julie Stephany Berrio, Mao Shan, and Stewart Worrall, “InterKey: Cross-modal Intersection Keypoints for Global Localization on OpenStreetMap”, arXiv preprint arXiv:2509.13857, 2025. Available here.

Citing

If you find this work useful please cite

@article{tran2025interkey,
  title   = {InterKey: Cross-modal Intersection Keypoints for Global Localization on OpenStreetMap},
  author  = {Tran, Nguyen Hoang Khoi and Berrio, Julie Stephany and Shan, Mao and Worrall, Stewart},
  journal = {arXiv preprint arXiv:2509.13857},
  year    = {2025}
}

Updated: