AI-Based Automated House Floor Plan Generation System Using Graph-Constrained GAN and Computational Geometry

Author

A. L. Narayana Reddy, B. Nikhila, B. Prasanna Kumar, D. Jaswanth, Dr. S. Akila Agnes

Abstract

  Designing a residential floor plan is slow and depends heavily on expert experience, while deep-learning generators usually output raster masks that cannot be measured, costed or visualised directly. This paper presents an AI-based automated house floor plan generation system that combines a pretrained graph-constrained generative model, HouseGAN++, with computational geometry and rule-based validation. From a plot and a list of required rooms, the system builds a room-relationship graph, generates several candidate layouts, repairs their geometry, scores them and selects a few distinct layouts. The chosen layout is stored as a single canonical description from which a furnished 2D plan, an interactive 3D model, a cost estimate and a PDF drawing are produced consistently. A sample run on a 40 ft × 60 ft plot shows that the system produces valid, measurable and furnished plans. The work shows that generative models become practical design tools when they are wrapped in graph, geometry and rule-based validation stages.

Keywords

Floor plan generation, HouseGAN++, Graph-conditioned GAN, Computational geometry, Binary Space Partitioning, Vastu Shastra, 3D visualisation, Bill of Quantities.

Full Text:

Download Paper PDF

References

[1] N. Nauata, S. Hosseini, K.-H. Chang, H. Chu, C.-Y. Cheng, and Y. Furukawa, “House-GAN++: Generative adversarial layout refinement network towards intelligent computational agent for professional architects,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), 2021.

[2] P. Merrell, E. Schkufza, and V. Koltun, “Computer-generated residential building layouts,” ACM Trans. Graph., vol. 29, no. 6, 2010.

[3] H. Fuchs, Z. M. Kedem, and B. F. Naylor, “On visible surface generation by a priori tree structures,” in Proc. ACM SIGGRAPH, 1980.

[4] M. de Berg, O. Cheong, M. van Kreveld, and M. Overmars, Computational Geometry: Algorithms and Applications, 3rd ed. Berlin, Germany: Springer, 2008.

[5] W. Wu, X.-M. Fu, R. Tang, Y. Wang, Y.-H. Qi, and L. Liu, “Data-driven interior plan generation for residential buildings,” ACM Trans. Graph., vol. 38, no. 6, 2019.

[6] R. Hu, Z. Huang, Y. Tang, O. van Kaick, H. Zhang, and H. Huang, “Graph2Plan: Learning floorplan generation from layout graphs,” ACM Trans. Graph., vol. 39, no. 4, 2020.

[7] N. Nauata, K.-H. Chang, C.-Y. Cheng, G. Mori, and Y. Furukawa, “House-GAN: Relational generative adversarial networks for graph-constrained house layout generation,” in Proc. Eur. Conf. Comput. Vis. (ECCV), 2020.

[8] F. Zhang, N. Nauata, and Y. Furukawa, “Conv-MPN: Convolutional message passing neural network for structured outdoor architecture reconstruction,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), 2020.

[9] N. Nauata, “houseganpp: House-GAN++ official implementation and pretrained checkpoint,” GitHub repository. [Online]. Available: https://github.com/ennauata/houseganpp

[10] C. Liu, J. Wu, P. Kohli, and Y. Furukawa, “Raster-to-vector: Revisiting floorplan transformation,” in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), 2017.

[11] R. T. Fielding, “Architectural styles and the design of network-based software architectures,” Ph.D. dissertation, Dept. Inf. Comput. Sci., Univ. California, Irvine, CA, USA, 2000.

[12] “Express: Fast, unopinionated, minimalist web framework for Node.js.” [Online]. Available: https://expressjs.com

[13] MongoDB, Inc., “MongoDB manual.” [Online]. Available: https://www.mongodb.com/docs/manual/

[14] S. Ramírez, “FastAPI framework, high performance, easy to learn, fast to code.” [Online]. Available: https://fastapi.tiangolo.com

[15] M. Jones, J. Bradley, and N. Sakimura, “JSON Web Token (JWT),” Internet Engineering Task Force, RFC 7519, May 2015.

[16] G. Bradski, “The OpenCV library,” Dr. Dobb’s Journal of Software Tools, 2000.

[17] A. A. Hagberg, D. A. Schult, and P. J. Swart, “Exploring network structure, dynamics, and function using NetworkX,” in Proc. 7th Python Sci. Conf. (SciPy), 2008, pp. 11–15.

[18] I. Goodfellow et al., “Generative adversarial nets,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 27, 2014.

[19] A. Paszke et al., “PyTorch: An imperative style, high-performance deep learning library,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 32, 2019.

[20] S. Gillies et al., “Shapely: Manipulation and analysis of geometric objects,” 2007–. [Online]. Available: https://github.com/shapely/shapely

[21] “Konva.js: HTML5 2D canvas library.” [Online]. Available: https://konvajs.org

[22] Meta Platforms, Inc., “React: The library for web and native user interfaces.” [Online]. Available: https://react.dev

[23] “three.js: JavaScript 3D library.” [Online]. Available: https://threejs.org

 

 

Share your valuable work from Social Media Buttons