LiDAR2BIM:

Speak the Same Language: Global LiDAR Registration on BIM Using Pose Hough Transform

Zhijian Qiao* , Haoming Huang*, Chuhao Liu, Zehuan Yu, Shaojie Shen, Fumin Zhang and Huan Yin

Accepted to IEEE Transactions on Automation Science and Engineering

*: Equal Contribution

News

• 03 Mar 2025: Accepted to [IEEE TASE].
• 01 May 2024: We released our paper on Arxiv and submit it to [IEEE TASE].

Abstract

LiDAR2BIM is a fast and robust global registration framework on BIM.

Features:

• Cross-modality Data Association: A comprehensive descriptor with O(1) retrieval complexity is proposed for cross-modality data association, effectively encoding basic walls and corners in structural environments.
• A Parallel and Robust Estimator based on Hierarchical Voting: A Hough transform scheme with hierarchical voting is introduced to hypothesize multiple pose candidates, followed by verifying the optimal transformation using an occupancy-aware score, improving both robustness and accuracy.

Note to Practitioners:

• Our proposed registration method lever- ages walls and corners as shared features between LiDAR and BIM data, making it particularly well-suited for scenarios with well-defined structural layouts. Accumulating a larger LiDAR submap provides richer structural information, which further aids in achieving accurate alignment. To optimize computational efficiency, we recommend constructing the descriptor database offline and loading it during runtime, enabling a theoretical retrieval complexity of O(1).
• Despite its advantages, our approach has certain limitations. First, it primarily focuses on planar structures, which limits its effectiveness in utilizing nonplanar features. Second, the method may underperform in cases where significant deviations exist between the as-designed BIM and as-is LiDAR data. Lastly, in ambiguous scenarios, such as long corridors or similar layouts within the same or across different floors, our method may struggle to verify the correct transformation among candidates. To address these challenges, incorporating additional information, particularly semantic cues such as floor numbers, room numbers, and room types, could enhance its robustness and reliability .

Getting Started

• Installation
• Data Preparation
• Demo
• Benchmarks

Qualitative results on datasets

Scene 1: Ground Floor Lobby in HKUST

scene_1.mp4

Scene 2: Office-3F in HKUST

scene_2.mp4

Scene 3: Office-4F in HKUST

scene_3.mp4

The LiBIM-UST dataset of this study has been upgraded to the SLABIM dataset, available at https://github.com/HKUST-Aerial-Robotics/SLABIM. For more comprehensive evaluation, please refer to the SLABIM dataset. (The tutorial of the SLABIM dataset will be released soon.)

Acknowledgements

We would like to show our greatest respect to authors of the following repos for making their works public:

• Teaser
• 3DBBS

Citation

If you find LiDAR2BIM or SLABIMdataset is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry.

@ARTICLE{10916778,
  author={Qiao, Zhijian and Huang, Haoming and Liu, Chuhao and Yu, Zehuan and Shen, Shaojie and Zhang, Fumin and Yin, Huan},
  journal={IEEE Transactions on Automation Science and Engineering}, 
  title={Speak the Same Language: Global LiDAR Registration on BIM Using Pose Hough Transform}, 
  year={2025},
  volume={},
  number={},
  pages={1-1},
  keywords={Point cloud compression;Robots;Laser radar;Three-dimensional displays;Buildings;Service robots;Semantics;Transforms;Sensors;Navigation;Point cloud registration;LiDAR;Building information modeling;Hough transform},
  doi={10.1109/TASE.2025.3549176}
  }

@article{huang2025slabim,
  title={SLABIM: A SLAM-BIM Coupled Dataset in HKUST Main Building},
  author={Huang, Haoming and Qiao, Zhijian and Yu, Zehuan and Liu, Chuhao and Shen, Shaojie and Zhang, Fumin and Yin, Huan},
  journal={arXiv preprint arXiv:2502.16856},
  year={2025}
}