TrueCity: Real and Simulated Urban Data for Cross-Domain 3D Scene Understanding
Duc Nguyen, Yan-Ling Lai, Qilin Zhang, Prabin Gyawali, Benedikt Schwab, Olaf Wysocki, Thomas H. Kolbe · 2025
Read TrueCity: Real and Simulated Urban Data for Cross-Domain 3D Scene Understanding by Duc Nguyen, Yan-Ling Lai, Qilin Zhang, Prabin Gyawali, Benedikt Schwab, Olaf Wysocki, Thomas H. Kolbe (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-07007.txt.
Description
Duc Nguyen and six co-authors introduce TrueCity, a benchmark for understanding three-dimensional city scenes from point clouds. Models need labeled real-world data to generalize, but labeling is scarce, and synthetic data, though perfectly labeled, misses real sensor noise and complexity. No earlier benchmark offered matched real and simulated point clouds of the same place for studying that synthetic-to-real gap.
TrueCity supplies centimeter-accurate labeled real point clouds, semantic 3D city models, and labeled simulated point clouds representing the same city. Its classes follow international 3D city modeling standards, enabling consistent comparison across the two domains. Extensive tests with common baselines quantify how large the domain shift is and test plans for using synthetic data to strengthen real-world scene understanding. Data, code, and models are public.
Catalog record
| CREATOR | Duc Nguyen, Yan-Ling Lai, Qilin Zhang, Prabin Gyawali, Benedikt Schwab, Olaf Wysocki, Thomas H. Kolbe |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-07007