OPAL: Visibility-aware LiDAR-to-OpenStreetMap Place Recognition via Adaptive Radial Fusion

Shuhao Kang, Youqi Liao, Yan Xia, Olaf Wysocki, Boris Jutzi, Daniel Cremers
Proceedings of The 9th Conference on Robot Learning, PMLR 305:4385-4400, 2025.

Abstract

LiDAR place recognition is a critical capability for autonomous navigation and cross-modal localization in large-scale outdoor environments. Existing approaches predominantly depend on pre-built 3D dense maps or aerial imagery, which impose significant storage overhead and lack real-time adaptability. In this paper, we propose OPAL, a novel network for LiDAR place recognition that leverages OpenStreetMap (OSM) as a lightweight and up-to-date prior. Our key innovation lies in bridging the domain disparity between sparse LiDAR scans and structured OSM data through two carefully designed components. First, a cross-modal visibility mask that identifies maximal observable regions from both modalities to guide feature learning. Second, an adaptive radial fusion module that dynamically consolidates radial features into discriminative global descriptors. Extensive experiments on the KITTI and KITTI-360 datasets demonstrate OPAL’s superiority, achieving 15.98% higher recall at @1m threshold for top-1 retrieved matches, along with 12x faster inference speed compared to the state-of-the-art approach. Code and datasets will be publicly available.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-kang25b, title = {OPAL: Visibility-aware LiDAR-to-OpenStreetMap Place Recognition via Adaptive Radial Fusion}, author = {Kang, Shuhao and Liao, Youqi and Xia, Yan and Wysocki, Olaf and Jutzi, Boris and Cremers, Daniel}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {4385--4400}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/kang25b/kang25b.pdf}, url = {https://proceedings.mlr.press/v305/kang25b.html}, abstract = {LiDAR place recognition is a critical capability for autonomous navigation and cross-modal localization in large-scale outdoor environments. Existing approaches predominantly depend on pre-built 3D dense maps or aerial imagery, which impose significant storage overhead and lack real-time adaptability. In this paper, we propose OPAL, a novel network for LiDAR place recognition that leverages OpenStreetMap (OSM) as a lightweight and up-to-date prior. Our key innovation lies in bridging the domain disparity between sparse LiDAR scans and structured OSM data through two carefully designed components. First, a cross-modal visibility mask that identifies maximal observable regions from both modalities to guide feature learning. Second, an adaptive radial fusion module that dynamically consolidates radial features into discriminative global descriptors. Extensive experiments on the KITTI and KITTI-360 datasets demonstrate OPAL’s superiority, achieving 15.98% higher recall at @1m threshold for top-1 retrieved matches, along with 12x faster inference speed compared to the state-of-the-art approach. Code and datasets will be publicly available.} }
Endnote
%0 Conference Paper %T OPAL: Visibility-aware LiDAR-to-OpenStreetMap Place Recognition via Adaptive Radial Fusion %A Shuhao Kang %A Youqi Liao %A Yan Xia %A Olaf Wysocki %A Boris Jutzi %A Daniel Cremers %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-kang25b %I PMLR %P 4385--4400 %U https://proceedings.mlr.press/v305/kang25b.html %V 305 %X LiDAR place recognition is a critical capability for autonomous navigation and cross-modal localization in large-scale outdoor environments. Existing approaches predominantly depend on pre-built 3D dense maps or aerial imagery, which impose significant storage overhead and lack real-time adaptability. In this paper, we propose OPAL, a novel network for LiDAR place recognition that leverages OpenStreetMap (OSM) as a lightweight and up-to-date prior. Our key innovation lies in bridging the domain disparity between sparse LiDAR scans and structured OSM data through two carefully designed components. First, a cross-modal visibility mask that identifies maximal observable regions from both modalities to guide feature learning. Second, an adaptive radial fusion module that dynamically consolidates radial features into discriminative global descriptors. Extensive experiments on the KITTI and KITTI-360 datasets demonstrate OPAL’s superiority, achieving 15.98% higher recall at @1m threshold for top-1 retrieved matches, along with 12x faster inference speed compared to the state-of-the-art approach. Code and datasets will be publicly available.
APA
Kang, S., Liao, Y., Xia, Y., Wysocki, O., Jutzi, B. & Cremers, D.. (2025). OPAL: Visibility-aware LiDAR-to-OpenStreetMap Place Recognition via Adaptive Radial Fusion. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:4385-4400 Available from https://proceedings.mlr.press/v305/kang25b.html.

Related Material