BuildingNet: Learning to Label 3D Buildings

TL;DR

BuildingNet provides a large-scale, annotated 3D building dataset and a graph neural network method that significantly improves labeling accuracy for complex building models, aiding various 3D vision and graphics tasks.

Contribution

The paper introduces BuildingNet, a comprehensive dataset of 3D building models with semantic labels, and a graph neural network approach for accurate mesh labeling, addressing the complexity of building structures.

Findings

The dataset contains 513K annotated mesh primitives across 2K models.

The proposed GNN method outperforms baseline models in mesh labeling accuracy.

BuildingNet enables advancements in 3D semantic segmentation and related tasks.

Abstract

We introduce BuildingNet: (a) a large-scale dataset of 3D building models whose exteriors are consistently labeled, (b) a graph neural network that labels building meshes by analyzing spatial and structural relations of their geometric primitives. To create our dataset, we used crowdsourcing combined with expert guidance, resulting in 513K annotated mesh primitives, grouped into 292K semantic part components across 2K building models. The dataset covers several building categories, such as houses, churches, skyscrapers, town halls, libraries, and castles. We include a benchmark for evaluating mesh and point cloud labeling. Buildings have more challenging structural complexity compared to objects in existing benchmarks (e.g., ShapeNet, PartNet), thus, we hope that our dataset can nurture the development of algorithms that are able to cope with such large-scale geometric data for both vision and graphics tasks e.g., 3D semantic segmentation, part-based generative models, correspondences, texturing, and analysis of point cloud data acquired from real-world buildings. Finally, we show that our mesh-based graph neural network significantly improves performance over several baselines for labeling 3D meshes.