This research introduces a three-dimensional (3D) building learning and generation framework based on graph theory and generative AI models, Building-Graph-AI. The framework aims to encode 3D building models into graph-structured data suitable for training graph neural networks (GNNs) and to generate layered 3D models at the detailed building component level. We test various encoding methods and neural networks, selecting the most effective method and defining it as Graph-BIM encoding. The results demonstrate that the Graph-BIM encoding method can reconstruct and generate detailed 3D building models from simple geometries and constraints. Compared to existing methods based on voxel, point cloud and 3D field, Building-Graph-AI excels in learning and generating detailed, hierarchical 3D models at the building component level, such as walls, columns and floors. By bridging the gap between geometric design and AI-based training and generation, this framework enhances the adaptability and efficiency of AI applications in architectural design.
EnjellinaBEVPRossyAGC. Review of AI image generator: influences, challenges, and future prospects for architectural field. J Artif Intell Archit2023; 2(1): 53–65.
2.
HegazyMSalehA. Evolution of AI role in architectural design: between parametric exploration and machine hallucination. MSA Engineering Journal2023; 2(2): 262–288.
3.
LiCZhangTDuX, et al.Generative AI models for different steps in architectural design: a literature review [Internet]. arXiv2024, [cited 2024 Nov 10]. Available from: https://arxiv.org/abs/2404.01335
4.
ChangKHChengCYLuoJ, et al.Building-GAN: graph-conditioned architectural volumetric design generation. In: 2021 IEEE/CVF int conf comput vis (ICCV), 2021. Preprint. DOI: 10.1109/iccv48922.2021.01174.
5.
ParkJJFlorencePStraubJ, et al.DeepSDF: learning continuous signed distance functions for shape representation [Internet]. arXiv2019, [cited 2025 Jan 31]. Available from: https://arxiv.org/abs/1901.05103
6.
ShenTGaoJYinK, et al.Deep marching tetrahedra: a hybrid representation for high-resolution 3D shape synthesis. Adv Neural Inf Process Syst2021; 34: 6087–6101.
7.
ZhongXKohIFrickerP. Building-GNN: exploring a co-design framework for generating controllable 3D building prototypes by graph and recurrent neural networks. In: Int Conf Educ Res Comput-Aided Archit Design Eur, 2023, pp. 431–440. eCAADe.
8.
HintonGESalakhutdinovRR. Reducing the dimensionality of data with neural networks. Science2006; 313(5786): 504–507.
9.
QiCRSuHMoK, et al.PointNet: deep learning on point sets for 3D classification and segmentation [Internet]. arXiv; 2017[cited 2025 Jan 31]. Available from: https://arxiv.org/abs/1612.00593
10.
HuangZWenYWangZ, et al.Surface reconstruction from point clouds: a survey and a benchmark [Internet]. arXiv; 2022[cited 2025 Jan 31]. Available from: https://arxiv.org/abs/2205.02413
11.
WuZSongSKhoslaA, et al.3D ShapeNets: a deep representation for volumetric shapes [Internet]. arXiv; 2015[cited 2025 Jan 31]. Available from: https://arxiv.org/abs/1406.5670
12.
MildenhallBSrinivasanPPTancikM, et al.NeRF: representing scenes as neural radiance fields for view synthesis. Commun ACM2022; 65(1): 99–106. DOI: 10.1145/3503250.
13.
XuKHuWLeskovecJ, et al.How powerful are graph neural networks?arXiv. Preprint arXiv:1810.008262018. DOI: 10.48550/ARXIV.1810.00826.
14.
SacksREastmanCMLeeG, et al.BIM handbook: a guide to building information modeling for owners, managers, designers, engineers and contractors. 3rd ed.Hoboken, New Jersey: Wiley, 2018, p. 659.
15.
NauataNChangKHChengCY, et al.Housegan: relational generative adversarial networks for graph-constrained house layout generation. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part I, Vol. 16, pp. 162–177.
16.
NauataNHosseiniSChangKH, et al.House-GAN++: generative adversarial layout refinement networks. 2021. [cited 2024 Apr 1]; Available from: https://arxiv.org/abs/2103.02574
17.
ZhaoPFeiYHuangY, et al.Design-condition-informed shear wall layout design based on graph neural networks. Adv Eng Inform2023; 58: 102190. DOI: 10.1016/j.aei.2023.102190.
18.
ZhaoPLiaoWHuangY, et al.Intelligent design of shear wall layout based on graph neural networks. Adv Eng Inform2023; 55: 101886.
BressonXLaurentT. A two-step graph convolutional decoder for molecule generation. arXiv. Preprint arXiv:1906.034122019.
21.
ChenJStouffsR. SE-VGAE: unsupervised disentangled representation learning for interpretable architectural layout design graph generation [Internet]. arXiv; 2024[cited 2025 Jan 31]. Available from: https://arxiv.org/abs/2406.17418
22.
ChenZTagliasacchiAZhangH. BSP-Net: generating compact meshes via binary space partitioning. In: Proc IEEE/CVF Conf Comput Vis Pattern Recognit, 2020, pp. 45–54.
23.
FangHLafargeFConnect-and-slice. A hybrid approach for reconstructing 3D objects. In: Proc IEEE/CVF Conf Comput Vis Pattern Recognit, 2020, pp. 13490–13498.
24.
HübnerPWeinmannMWursthornS, et al.Automatic voxel-based 3D indoor reconstruction and room partitioning from triangle meshes. ISPRS J Photogramm Remote Sens2021; 181: 254–278.
25.
ChangKHChengCY. Learning to simulate and design for structural engineering. arXiv (Cornell Univ). Preprint2020. https://arxiv.org/pdf/2003.09103.pdf
26.
SimonovskyMKomodakisN. GraphVAE: towards generation of small graphs using variational autoencoders. In: Artif Neural Netw Mach Learn–ICANN 2018: 27th Int Conf Artif Neural Netw. Rhodes, Greece, 4–7 October 2018, pp. 412–422.
27.
WinterRNoéFClevertDA. Permutation-invariant variational autoencoder for graph-level representation learning. Adv Neural Inf Process Syst2021; 34: 9559–9573. https://arxiv.org/abs/2104.09856
28.
ShiYHuangZFengS, et al.Masked label prediction: unified message passing model for semi-supervised classification. arXiv. Preprint arXiv:2009.035092020. DOI: 10.48550/ARXIV.2009.03509.
29.
KipfTNWellingM. Semi-supervised classification with graph convolutional networks. arXiv. Preprint arXiv:1609.029072016.
30.
HamiltonWLYingRLeskovecJ. Inductive representation learning on large graphs. arXiv (Cornell Univ). Preprint2017. https://arxiv.org/pdf/1706.02216.pdf
VaswaniAShazeerNParmarN, et al.Attention is all you need [internet]. arXiv2023, [cited 2025 Jan 31]. Available from: https://arxiv.org/abs/1706.03762
33.
SchusterMPaliwalKK. Bidirectional recurrent neural networks. IEEE Trans Signal Process1997; 45(11): 2673–2681.
34.
YuYSiXHuC, et al.A review of recurrent neural networks: LSTM cells and network architectures. Neural Comput2019; 31(7): 1235–1270.
35.
MeschederLOechsleMNiemeyerM, et al.Occupancy networks: learning 3D reconstruction in function space [Internet]. arXiv2019[cited 2025 Jan 31]. Available from: https://arxiv.org/abs/1812.03828
