Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BuildingWorld: A Structured 3D Building Dataset for Urban Foundation Models

Shangfeng Huang, Ruisheng Wang, Xin Wang

TL;DR

BuildingWorld introduces a globally diverse LoD2 3D building dataset comprising roughly 5 million building meshes from 44 cities, paired with real and simulated airborne LiDAR data. It adds Cyber City, a procedural urban scene generator, and uses Helios++ to simulate realistic ALS point clouds, enabling scalable training for urban foundation modeling. A benchmark shows synthetic data can generalize to real §point clouds, highlighting the dataset’s potential to bridge the synthetic-to-real gap for 3D reconstruction and analysis in digital twins. The work also provides LoD3 extensions and semantic/instance annotations to support advanced 3D understanding tasks in diverse urban contexts.

Abstract

As digital twins become central to the transformation of modern cities, accurate and structured 3D building models emerge as a key enabler of high-fidelity, updatable urban representations. These models underpin diverse applications including energy modeling, urban planning, autonomous navigation, and real-time reasoning. Despite recent advances in 3D urban modeling, most learning-based models are trained on building datasets with limited architectural diversity, which significantly undermines their generalizability across heterogeneous urban environments. To address this limitation, we present BuildingWorld, a comprehensive and structured 3D building dataset designed to bridge the gap in stylistic diversity. It encompasses buildings from geographically and architecturally diverse regions -- including North America, Europe, Asia, Africa, and Oceania -- offering a globally representative dataset for urban-scale foundation modeling and analysis. Specifically, BuildingWorld provides about five million LOD2 building models collected from diverse sources, accompanied by real and simulated airborne LiDAR point clouds. This enables comprehensive research on 3D building reconstruction, detection and segmentation. Cyber City, a virtual city model, is introduced to enable the generation of unlimited training data with customized and structurally diverse point cloud distributions. Furthermore, we provide standardized evaluation metrics tailored for building reconstruction, aiming to facilitate the training, evaluation, and comparison of large-scale vision models and foundation models in structured 3D urban environments.

BuildingWorld: A Structured 3D Building Dataset for Urban Foundation Models

TL;DR

BuildingWorld introduces a globally diverse LoD2 3D building dataset comprising roughly 5 million building meshes from 44 cities, paired with real and simulated airborne LiDAR data. It adds Cyber City, a procedural urban scene generator, and uses Helios++ to simulate realistic ALS point clouds, enabling scalable training for urban foundation modeling. A benchmark shows synthetic data can generalize to real §point clouds, highlighting the dataset’s potential to bridge the synthetic-to-real gap for 3D reconstruction and analysis in digital twins. The work also provides LoD3 extensions and semantic/instance annotations to support advanced 3D understanding tasks in diverse urban contexts.

Abstract

As digital twins become central to the transformation of modern cities, accurate and structured 3D building models emerge as a key enabler of high-fidelity, updatable urban representations. These models underpin diverse applications including energy modeling, urban planning, autonomous navigation, and real-time reasoning. Despite recent advances in 3D urban modeling, most learning-based models are trained on building datasets with limited architectural diversity, which significantly undermines their generalizability across heterogeneous urban environments. To address this limitation, we present BuildingWorld, a comprehensive and structured 3D building dataset designed to bridge the gap in stylistic diversity. It encompasses buildings from geographically and architecturally diverse regions -- including North America, Europe, Asia, Africa, and Oceania -- offering a globally representative dataset for urban-scale foundation modeling and analysis. Specifically, BuildingWorld provides about five million LOD2 building models collected from diverse sources, accompanied by real and simulated airborne LiDAR point clouds. This enables comprehensive research on 3D building reconstruction, detection and segmentation. Cyber City, a virtual city model, is introduced to enable the generation of unlimited training data with customized and structurally diverse point cloud distributions. Furthermore, we provide standardized evaluation metrics tailored for building reconstruction, aiming to facilitate the training, evaluation, and comparison of large-scale vision models and foundation models in structured 3D urban environments.

Paper Structure

This paper contains 17 sections, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Related Datasets
  4. Related Methods
  5. BuildingWorld Dataset
  6. Overview
  7. Building Models
  8. Simulated ALS
  9. Cyber City
  10. Benchmark
  11. Baseline Methods
  12. Datasets and Experimental Setup
  13. Results and Discussion
  14. Challenges and Tasks
  15. LoD3 Building Reconstruction
  16. ...and 2 more sections

Figures (4)

  • Figure 1: A glimpse of the BuildingWorld dataset.
  • Figure 2: Illustration of the construction process of BuildingWorld dataset. A glimpse of the LOD2 digital city model of Boston is shown. The zoomed-in downtown area illustrates simulated aerial LiDAR point clouds, generated using a predefined airborne platform, LiDAR sensor, and flight trajectory.
  • Figure 3: Statistical overview of the BuildingWorld dataset. Area bars indicate scene sizes, while percentage area and height metrics highlight the diversity of building structures.
  • Figure 4: Illustration of Cyber City, which consists of four main components: terrain, road and building footprints, buildings, and vegetation.