This repository contains the foundational open-source research version of the Cloud2BIM algorithm. An advanced, production-ready version of this software, Cloud2BIM-AI, is now available through Constriq (a Czech Technical University spin-off).
While the open-source script provides core functionality, Cloud2BIM-AI is actively developed for commercial and enterprise workflows, offering the following expanded capabilities:
- Enhanced Element Recognition: Advanced AI models capable of recognizing a wider variety of complex building elements with higher precision.
- Increased Robustness: Fully automated processing that eliminates the need for manual parameter tuning.
- Optimized Performance: Significantly reduced computational time for large-scale point cloud datasets.
- Continuous Support & GUI: A user-friendly interface with visual verification and ongoing technical development.
For professional deployment and access to the advanced version, please visit: constriq.tech
Cloud2BIM automates the Scan-to-BIM process by converting point clouds into 3D parametric entities. It employs a segmentation algorithm that utilizes point cloud density analysis, augmented by image and morphological operations. This allows the software to precisely extract the geometry of building elements such as slabs, walls, windows, and doors.
To install Cloud2BIM, follow these steps: git clone
https://github.com/VaclavNezerka/Cloud2BIM.git
Install dependencies:
First, ensure you have Python and pip installed. Then, install the required dependencies listed in the requirements.txt file:
pip install -r requirements.txt
The cloud2entities.py script requires a YAML configuration file to run. You can provide the path to the
configuration file as a command-line argument. If no argument is provided, the script will
automatically use the default file config.yaml.
python cloud2entities.py config.yaml
The complete original point cloud for Kladno station is available at Zenodo platform.
https://zenodo.org/records/14221915
If you find this project or any part of it useful in your research or work, please consider citing the following article:
@article{Cloud2BIM_2025, title = {Open-source automatic pipeline for efficient conversion of large-scale point clouds to IFC format}, journal = {Automation in Construction}, volume = {177}, pages = {106303}, year = {2025}, issn = {0926-5805}, doi = {https://doi.org/10.1016/j.autcon.2025.106303}, author = {Slávek Zbirovský and Václav Nežerka}, }
MIT License