A generative artificial intelligence approach to modular skeletal framework modeling: Bamboo stilt houses as a case study

Abstract

This paper presents a new generative artificial intelligence (AI) approach for creating modular skeletal frameworks, using vernacular bamboo stilt houses as examples to investigate an innovative methodological perspective. By transforming building skeletons to connected graphs, our method uses Variational Graph Autoencoders (VGAE) and Graph Sample and Aggregate (GraphSAGE) to generate 3D modular components based on spatial constraints set by users, such as axis grids and chosen room areas. The graph representation encodes structural elements as edges and their connections as nodes, maintaining critical dimensional constraints and spatial relationships. Using data from bamboo stilt houses built without architects, we make a specialized dataset of geometric skeletons for model training. Experimental results demonstrate the effectiveness of our approach in capturing the distribution of featured elements in building frameworks and in generating structurally sound designs, with GraphSAGE showing better performance compared to alternative methods. The probabilistic edge prediction approach allows for a collaborative human-AI design process, empowering designers while utilizing computational capabilities. The inherent flexibility of the graph-based representation makes it adaptable to a wide range of materials and scales.

Keywords

Graph neural networks / Modular components / Bamboo architecture / Computational design / Generative artificial intelligence method

References

[1] Alymani, A. , Jabi, W. , 2024. A graph-based computational tool for retrieving architectural precedents of building and ground relationship (BGR tool). Int. J. Architect. Comput. 0, 14780771241260853.

[2] Alymani, A. , Jabi, W. , Corcoran, P. , 2023. Graph machine learning classification using architectural 3D topological models. Simulation 99 (11), 1117- 1131.

[3] Bleker, L. , Pastrana, R. , Ohlbrock, P.O. , D'Acunto, P. , 2023. Structural form-finding enhanced by graph neural networks. In: Gengnagel, C., Baverel, O., Betti, G., Popescu, M., Thomsen, M. R., Wurm, J. (Eds.), Towards Radical Regeneration. Springer, pp. 24-35.

[4] Bleker, L. , Tam, K.-M.M. , D'Acunto, P. , 2024. Logic-informed graph neural networks for structural form-finding. Adv. Eng. Inform. 61, 102510.

[5] Chang, A.X. , Funkhouser, T. , Guibas, L. , Hanrahan, P. , Huang, Q. , Li, Z. , Savarese, S. , Savva, M. , Song, S. , Su, H. , Xiao, J. , 2015. Shapenet: an information-rich 3d model repository. arXiv preprint, arXiv: 1512.03012.

[6] Chang, K.-H. , Cheng, C.-Y. , Luo, J. , Murata, S. , Nourbakhsh, M. , Tsuji, Y. , 2021. Building-GAN: graph-conditioned architectural volumetric design generation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 11956-11965.

[7] de Miguel Rodríguez, J. , Villafanẽ M.E. , Piškorec, L. , Caparrini, F. S. , 2020. Generation of geometric interpolations of building types with deep variational autoencoders. Des. Sci. 6, e34.

[8] Di Carlo, R. , Mittal, D. , Veselý O. , 2022. Generating 3D building volumes for a given urban context using Pix2Pix GAN. In: Proceedings of the 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, pp. 287-295.

[9] Ennemoser, B. , Mayrhofer-Hufnagl, I. , 2023. Design across multiscale datasets by developing a novel approach to 3DGANs. Int. J. Architect. Comput. 21 (2), 358- 373.

[10] Goodfellow, I. , Pouget-Abadie, J. , Mirza, M. , Xu, B. , WardeFarley, D. , Ozair, S. , Courville, A. , Bengio, Y. , 2014. Generative adversarial nets. Adv. Neural Inf. Process. Syst. 27, 2672- 2680.

[11] Gori, M. , Monfardini, G. , Scarselli, F. , 2005. A new model for learning in graph domains. In: Proceedings of 2005 IEEE International Joint Conference on Neural Networks. IEEE, pp. 729-734.

[12] Hamilton, W. , Ying, Z. , Leskovec, J. , 2017. Inductive representation learning on large graphs. Adv. Neural Inf. Process. Syst. 30, 1025- 1035.

[13] Hayashi, K. , Ohsaki, M. , 2022. Graph-based reinforcement learning for discrete cross-section optimization of planar steel frames. Adv. Eng. Inform. 51, 101512.

[14] Horvath, A.S. , Pouliou, P. , 2024. AI for conceptual architecture: reflections on designing with text-to-text, text-to-image, and image-to-image generators. Front. Architect. Res. 13 (3), 593- 612.

[15] Hua, H. , Hovestadt, L. , Wang, Q. , 2024. Flexible high-rise apartments with sparse wall-frame structure: a data-driven computational approach. Front. Architect. Res. 13 (3), 639- 649.

[16] Kim, D. , Lee, L.S. , Kim, H. , 2023. Elemental sabotage: diffusing functional morphologies. In: P Proceedings of the 28th Conference on Computer Aided Architectural Design Research in Asia(CAADRIA 2023), pp. 29-38.

[17] Kim, F.C. , Huang, J. , 2022. Deep architectural archiving (DAA), towards a machine understanding of architectural form. In: Proceedings of the 27th Conference on Computer Aided Architectural Design Research in Asia (CAADRIA 2022), vol. 1, pp. 727-736.

[18] Kingma, D.P. , Welling, M. , 2013. Auto-encoding variational bayes. arXiv preprint, arXiv: 1312.6114.

[19] Kipf, T.N. , Welling, M. , 2016a. Semi-supervised classification with graph convolutional networks. arXiv preprint, arXiv: 1609.02907.

[20] Kipf, T.N. , Welling, M. , 2016b. Variational graph auto-encoders. arXiv preprint, arXiv: 1611.07308.

[21] Lala, S. , Gopalakrishnan, N. , Kumar, A. , 2017. A comparative study on the seismic performance of the different types of bamboo stilt houses of North-East India. J. Environ. Nanotechnol. 6 (2), 59- 73.

[22] Li, C. , Zhang, T. , Du, X. , Zhang, Y. , Xie, H. , 2025. Generative AI models for different steps in architectural design: a literature review. Front. Architect. Res. 14, 759- 783.

[23] Li, X. , Zhang, Q. , Kang, D. , Cheng, W. , Gao, Y. , Zhang, J. , Liang, Z. , Liao, J. , Cao, Y.-P. , Shan, Y. , 2024. Advances in 3D generation: a survey. arXiv preprint, arXiv: 2401.17807.

[24] Liang, J. , Zhong, X. , Koh, I. , 2024. Bridging BIM and AI: a graph-BIM encoding approach for detailed 3D layout generation using variational graph autoencoder. In: Proceedings of the 29th Conference on Computer-Aided Architectural Design Research in Asia (CAADRIA 2024), pp. 221-230.

[25] Liu, K. , Jayaraman, D. , Estrella, P. , Shi, Y. , Yang, J. , Wu, J. , Lopez, L. , Escardó, B. , Moreno, F. , 2022. Create an enabling environment for bamboo construction sector in Africa, Asia and Latin America. In: International Conference on Nonconventional Materials and Technologies (Virtual Conference).

[26] Liu, W. , Zang, Y. , Xiong, Z. , Bian, X. , Wen, C. , Lu, X. , Wang, C. , Marcato, J. , Gonçalves, W.N. , Li, J. , 2023. 3D building model generation from MLS point cloud and 3D mesh using multisource data fusion. Int. J. Appl. Earth Obs. Geoinf. 116, 103171.

[27] Nash, C. , Ganin, Y. , Eslami, S.A. , Battaglia, P. , 2020. Polygen: an autoregressive generative model of 3d meshes. In: Proceedings of the 37th International Conference on Machine Learning, pp. 7220-7229.

[28] Paudel, S.K. , Lobovikov, M. , 2003. Bamboo housing: market potential for low-income groups. J. Bamboo Rattan 2 (4), 381- 396.

[29] Poole, B. , Jain, A. , Barron, J.T. , Mildenhall, B. , 2022. DreamFusion: text-to-3D using 2D diffusion. arXiv preprint, arXiv: 2209.14988.

[30] Qi, C.R. , Su, H. , Mo, K. , Guibas, L.J. , 2017. PointNet: deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 652-660.

[31] Ren, X. , Huang, J. , Zeng, X. , Museth, K. , Fidler, S. , Williams, F. , 2024. X-Cube: large-scale 3D generative modeling using sparse voxel hierarchies. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4209-4219.

[32] Ren, Y. , Zheng, H. , 2020. The spire of AI: voxel-based 3D neural style transfer. In: Proceedings of the 25th International Conference on Computer-Aided Architectural Design Research in Asia (CAADRIA 2020), pp. 619-628.

[33] Saiful, N. , Kadir, T.A.Q.R.A. , 2022. Sustainable on-stilt construction technology for mangrove land in Malaysia. MAJ-Malaysia Architect. J. 4 (3), 43- 50.

[34] Scarselli, F. , Gori, M. , Tsoi, A.C. , Hagenbuchner, M. , Monfardini, G. , 2009. The graph neural network model. IEEE Trans. Neural Network. 20 (1), 61- 80.

[35] Sebestyen, A. , Hirschberg, U. , Rasoulzadeh, S. , 2023a. Using deep learning to generate design spaces for architecture. Int. J. Architect. Comput. 21 (4), 337- 357.

[36] Sebestyen, A. , Özdenizci, O. , Legenstein, R. , Hirschberg, U. , 2023b. Generating conceptual architectural 3D geometries with denoising diffusion models. In: Proceedings of the 41st International Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe), 2, pp. 451-460.

[37] Shim, J. , Kang, C. , Joo, K. , 2023. Diffusion-based signed distance fields for 3D shape generation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 20887-20897.

[38] Song, Y. , Ermon, S. , 2019. Generative modeling by estimating gradients of the data distribution. Adv. Neural Inf. Process. Syst. 32, 11886- 11898.

[39] Tam, K.-M.M. , Van Mele, T. , Block, P. , 2022. Trans-topological learning and optimisation of reticulated equilibrium shell structures with automatic differentiation and CW complexes message passing. In: Proceedings of IASS Annual Symposia, pp. 1-13.

[40] Vishwanath, K.V. , Gupta, D. , Vahdat, A. , Yocum, K. , 2009. ModelNet: towards a datacenter emulation environment. In: 2009 IEEE Ninth International Conference on peer-to-peer Computing, pp. 81-82.

[41] Whalen, E. , Mueller, C. , 2021. Toward reusable surrogate models: graph-based transfer learning on trusses. J. Mech. Des. 144 (2), 021704.

[42] Wu, W. , Fu, X.-M. , Tang, R. , Wang, Y. , Qi, Y.-H. , Liu, L. , 2019. Data-driven interior plan generation for residential buildings. ACM Trans. Graph. 38 (6), 1-12. Article 234.

[43] Zhang, C. , Tao, M.-X. , Wang, C. , Fan, J.-S. , 2024. End-to-end generation of structural topology for complex architectural layouts with graph neural networks. Comput. Aided Civ. Infra-struct. Eng. 39 (5), 756- 775.

[44] Zhang, J. , Li, X. , Wan, Z. , Wang, C. , Liao, J. , 2024. Text2NeRF: text-driven 3D scene generation with neural radiance fields. IEEE Trans. Visual. Comput. Graph. 30 (12), 7749- 7762.

[45] Zhao, P. , Liao, W. , Huang, Y. , Lu, X. , 2024. Beam layout design of shear wall structures based on graph neural networks. Autom. ConStruct. 158, 105223.

[46] Zheng, H. , Yuan, P.F. , 2021. A generative architectural and urban design method through artificial neural networks. Build. Environ. 205, 108178.

[47] Zhong, X. , Koh, I. , Fricker, P. , 2023. Building-GNN: exploring a co-design framework for generating controllable 3D building prototypes by graph and recurrent neural networks. In: Proceedings of the 42st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023), pp. 431-440.

[48] Zhong, X. , Liang, J. , Li, Y. , 2024. Building-agent: a 3D generation agent framework integrating large language models and graph-based 3D generation model. In: Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2024), pp. 291-300.

[49] Zhou, Y. , Leng, H. , Meng, S. , Wu, H. , Zhang, Z. , 2024. StructDiffusion: end-to-end intelligent shear wall structure layout generation and analysis using diffusion model. Eng. Struct. 309, 118068.

[50] Zhuang, J. , Li, G. , Xu, H. , Xu, J. , Tian, R. , 2024. Text-to-City: controllable 3D urban block generation with latent diffusion model. In: Proceedings of the 29th International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA 2024), pp. 169-178.

[51] Zhuang, X. , Ju, Y. , Yang, A. , Caldas, L. , 2023. Synthesis and generation for 3D architecture volume with generative modeling. Int. J. Architect. Comput. 21 (2), 297- 314.

RIGHTS & PERMISSIONS

The Author(s). Publishing services by Elsevier B.V. on behalf of Higher Education Press and KeAi.