SemanticBridge - A Dataset for 3D Semantic Segmentation of Bridges and Domain Gap Analysis

📖 Paper | 🛠️ Github

⭐️ Overview

This dataset contains large-scale 3D point clouds of bridges, captured using different sensors. This enables investigation into different approaches for large object segmentation and 3D domain gap analysis. Furthermore, it enables the study of bridge structures, addressing a critical need in infrastructure inspection and maintenance. The dataset consists of a total of 20 real-world bridges, ranging in length from 15 to 70 meters, which have been annotated into nine different classes. Some of the bridges were scanned with multiple devices to enable domain gap analysis. The dataset can be enriched by adding 23 synthetic bridges that share the same classes.

👋 Citation

If you find our paper, data or code useful in your research, please cite our paper.

@article{SemanticBridge,
title = {SemanticBridge—A dataset for 3D semantic segmentation of bridges and domain gap analysis},
journal = {Developments in the Built Environment},
volume = {26},
pages = {100912},
year = {2026},
issn = {2666-1659},
doi = {https://doi.org/10.1016/j.dibe.2026.100912},             
url = {https://www.sciencedirect.com/science/article/pii/S2666165926000700},
author = {Maximilian Kellner and Mariana {Ferrandon Cervantes} and Yuandong Pan and Ruodan Lu and Ioannis Brilakis and Alexander Reiterer},
}