A dual-aspect evaluation framework for architectural-like plan generation via pix2pix series algorithms

Abstract

Architectural plan generation via pix2pix series algorithms faces dual challenges: the absence of domain-specific evaluation metrics and a lack of systematic insights into the joint impact of training configurations. To address the limitations of pix2pix-based models adaptation to architectural design, we designed a training regimen involving 12 experiments with varying training set sizes, dataset characteristics, and algorithms. These experiments utilized our self-built, high-quality, large-volume synthetic dataset of architectural-like plans. By saving intermediate models, we obtained 240 generative models for evaluation on a fixed test set. To quantify model performance, we developed a dual-aspect evaluation method that assesses predictions through pixel similarity (principle adherence) and segmentation line continuity (vectorization quality). Analysis revealed algorithm choice and training set size as primary factors, with larger sets enhancing the benefits of high-resolution and enhanced-annotation datasets. The optimal model achieved high-quality predictions, demonstrating strict adherence to predefined principles (0.81 similarity) and effective vectorization (0.86 segmentation line continuity). Testing on 7695 samples of varying complexity confirmed the model's robustness, strong generative capability, and controlled innovation within defined principles, validated through 3D model conversion. This work provides a domain-adapted framework for training and evaluating pix2pix-based architectural generators, bridging generative research and practical applications.

Keywords

Architectural-like plan generation / Quantitative evaluation / pix2pix / pix2pixHD / Vectorization / Deep learning

References

[1] Alotaibi, A. , 2020. Deep generative adversarial networks for imageto-image translation: a review. Symmetry 12 (10), 1705.

[2] Arjovsky, M. , Bottou, L. , 2017. Towards principled methods for training generative adversarial networks. arXiv: 1701.04862.

[3] As, I. , Pal, S. , Basu, P. , 2018. Artificial intelligence in architecture: generating conceptual design via deep learning. Int. J. Architect. Comput. 16 (3), 306- 327.

[4] Borji, A. , 2019. Pros and cons of GAN evaluation measures. Comput. Vis. Image Understand. 179, 41- 65.

[5] Cao, Y. , Jia, L. , Chen, Y. , Lin, N. , Li, X. , 2018. Review of computer vision based on generative adversarial networks. J. Image Graph. 23 (12), 1433-1449 (in Chinese).

[6] Chaillou, S. , 2021. AI & architecture: an experimental perspective(Degree of Master). In: The Routledge Companion to Artificial Intelligence in Architecture. Routledge.

[7] Chakraborty, T. , S, U.R.K. , Naik, S.M. , Panja, M. , Manvitha, B. , 2024. Ten years of generative adversarial nets (GANs): a survey of the state-of-the-art. Mach. Learn. Sci. Technol. 5 (1), 011001.

[8] Charman, P. , 1994. A constraint-based approach for the generation of floor plans. In: Proceedings Sixth International Conference on Tools with Artificial Intelligence. TAI 94. IEEE, New Orleans, Los Angeles, USA, pp. 555-561.

[9] Chen, A. , 2022. Generation of layouts for living spaces using conditional generative adversarial networks. In: Designing Floor Plans that Respect Both a Boundary and high-level Requirements (Degree of Master). KTH Royal Institute of Technology, Stockholm.

[10] Chen, M. , Zheng, H. , Wu, J. , 2022. Computational design of multi-functional system based on generative adversarial networks taking the layout generation of vocational and technical college as an example. Archit. J. S1, 103-108 (in Chinese).

[11] Coorey, B.P. , Jupp, J.R. , 2014. Generative spatial performance design system. AI-EDAM 28 (3), 277- 283.

[12] Cui, Z. , Guo, Y. , Ishimoto, T. , 2020. Research on AI aided nursing home architectural-design based on cGANs. Hous. Sci. 40 (1), 31-34 (in Chinese).

[13] Cui, Z. , Mori, T. , Li, H. , 2023. Definition of digital built environment in the development of digital transformation of architecture. World Architect. 7, 38- 39.

[14] Davis, R.A. , Nielsen, M.S. , 2020. Modeling of time series using random forests: theoretical developments. Electron. J. Stat. 14 (2), 3644- 3671.

[15] Dino, I.G. , 2016. An evolutionary approach for 3D architectural space layout design exploration. Autom. ConStruct. 69, 131- 150.

[16] Egor, G. , Sven, S. , Martin, D. , Reinhard, K. , 2020. Computer-aided approach to public buildings floor plan generation. Magnetizing floor plan generator. Procedia Manuf. 44, 132- 139.

[17] Eisenstadt, V. , Langenhan, C. , Althof, K.-D. , 2019. Generation of floor plan variations with convolutional neural networks and case-based reasoning: an approach for transformative adaptation of room configurations within a framework for support of early conceptual design phases. In: Proceedings of the 37th Education and Research in Computer Aided Architectural Design in Europe(eCAADe) and 23rd Iberoamerican Society of Digital Graphics(SIGraDi), Joint Conference (N. 1), Vol. 2. Brussels, pp. 79-84.

[18] Flemming, U. , Coyne, R. , Glavin, T. , Rychener, M. , 1988. A generative expert system for the design of building layouts: version 2. Carnegie Mellon University, Engineering Design Research Center.

[19] Genuer, R. , Poggi, J.-M. , Tuleau-Malot, C. , 2010. Variable selection using random forests. Pattern Recognit. Lett. 31 (14), 2225- 2236.

[20] Gero, J.S. , Kazakov, V.A. , 1998. Evolving design genes in space layout planning problems. Artif. Intell. Eng. 12 (3), 163- 176.

[21] Guo, Y. , Cui, Z. , 2024. Impact of the training set consistency on architectural plan generation effect based on Pix2pixHD algorithm. In: Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe(eCAADe 2024), Vol. 1, pp. 459-468. Nicosia, Cyprus.

[22] Heusel, M. , Ramsauer, H. , Unterthiner, T. , Nessler, B. , Hochreiter, S. , 2017. GANs trained by a two time-scale update rule converge to a local nash equilibrium. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. NIPS'17. Curran Associates Inc., Red Hook, New York, USA, pp. 6629-6640.

[23] Hu, R. , Huang, Z. , Tang, Y. , Van Kaick, O. , Zhang, H. , Huang, H. , 2020. Graph2Plan: learning floorplan generation from layout graphs. ACM Trans. Graph. 39 (4), 118: 1- 118: 14.

[24] Huang, W. , Zheng, H. , 2018. Architectural drawings recognition and generation through machine learning. In: Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA), pp. 156-165. Mexico City, Mexico.

[25] Hayes-Roth, F. , Waterman, D.A. , Lenat, D.B. , 1983. Building Expert Systems. Addison-Wesley Publishing Company, Massachusetts, USA, pp. 3-30.

[26] Isola, P. , Zhu, J.-Y. , Zhou, T. , Efros, A.A. , 2017. Image-to-Image translation with conditional adversarial networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). IEEE, Honolulu, HI, pp. 5967-5976.

[27] Jabbar, A. , Li, X. , Omar, B. , 2021. A survey on generative adversarial networks: variants, applications, and training. ACM Comput. Surv. 54 (8), 1- 49.

[28] Jin, L. , Tan, F. , Jiang, S. , 2020. Generative adversarial network technologies and applications in computer vision. Comput.Intell. Neurosci. 1, 459107.

[29] Karadag, I. , Güzelci, O.Z. , Alaçam, S. , 2022. EDU-AI: a twofold machine learning model to support classroom layout generation. Constr. Innov. 23 (4), 898- 914.

[30] Kim, T. , Cha, M. , Kim, H. , Lee, J.K. , Kim, J. , 2017. Learning to discover cross-domain relations with generative adversarial networks. Proceedings of the 34th International Conference on Machine Learning (ICML) 70, 1857- 1865.

[31] Laignel, G. , Pozin, N. , Geffrier, X. , Delevaux, L. , Brun, F. , Dolla, B. , 2021. Floor plan generation through a mixed constraint programming-genetic optimization approach. Autom. ConStruct. 123, 103491.

[32] Liu, D. , Wang, K. , 2022. Research on the generative design of architectural form and combination based on Pix2pix algorithm. Build. Sci. 38 (3), 260- 267+286.

[33] Medjdoub, B. , Yannou, B. , 2000. Separating topology and geometry in space planning. Comput. Aided Des. 32 (1), 39- 61.

[34] Merrell, P. Schkufza, E. Koltun, V. , 2010. Computer-generated residential building layouts. In: ACM SIGGRAPH Asia 2010 Papers. ACM Press, Seoul, South Korea, pp. 181: 1-181: 12.

[35] Mirza, M. , Osindero, S. , 2014. Conditional generative adversarial nets. arXiv: 1411.1784v1[cs.LG].

[36] Mitchell, W.J. , Steadman, J.P. , Liggett, R.S. , 1976. Synthesis and optimization of small rectangular floor plans. Environ. Plann. Plann. Des. 3 (1), 37- 70.

[37] Naqbi, H. , Bahroun, Z. , Ahmed, V. , 2024. Enhancing work productivity through generative artificial intelligence: a comprehensive literature review. Sustainability 16 (2), 1166.

[38] Nauata, N. , Chang, K.-H. , Cheng, C.-Y. , Mori, G. , Furukawa, Y. , 2020. House-GAN: relational generative adversarial networks for graph-constrained house layout generation. In: Papers of European Conference on Computer Vision 2020. Springer International Publishing, Glasgow, UK, pp. 162-177.

[39] Negroponte, N. , 2015. Being Digital. Knopf Doubleday Publishing Group, New York.

[40] Qian, W. , Xu, Y. , Li, H. , 2022. A self-sparse generative adversarial network for autonomous early-stage design of architectural sketches. Comput. Aided Civil. Eng. 37 (7), 612- 628.

[41] Sato, G. , Ishizawa, T. , Iseda, H. , Kitahara, H. , 2020. Automatic generation of the schematic mechanical system drawing by generative adversarial network. In: Proceedings of the 38th Education and Research in Computer Aided Architectural Design in Europe (eCAADe) Education and Research in Computer Aided Architectural Design in Europe (eCAADe), pp. 403-410. Berlin, Germany.

[42] Schwarz, A. , Berry, D.M. , Shaviv, E. , 1994a. On the use of the automated building design system. Comput. Aided Des. 26 (10), 747- 762.

[43] Schwarz, A. , Berry, D.M. , Shaviv, E. , 1994b. Representing and solving the automated building design problem. Comput. Aided Des. 26 (9), 689- 698.

[44] Sharma, D. , Gupta, N. , Chattopadhyay, C. , Mehta, S. , 2018. REXplore: a sketch based interactive explorer for real estates using building floor plan images. In: Proceedings of 2018 IEEE International Symposium on Multimedia (ISM). IEEE Computer Society, pp. 61-64.

[45] Sun, C. , Hu, W. , 2020. A rapid building density survey method based on improved unet. In: Proceedings of the 25th ComputerAided Architectural Design Research in Asia Conference (CAADRIA), vol. 2. Chulalongkorn University, Bangkok, Thailand, pp. 649-658.

[46] Taigman, Y. , Polyak, A. , Wolf, L. , 2016. Unsupervised cross-domain image generation. arXiv: 1611.02200[cs.CV].

[47] Wang, Z. , Bovik, A.C. , Sheikh, H.R. , Simoncelli, E.P. , 2004. Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13 (4), 600- 612.

[48] Wang, S. , Zeng, W. , Chen, X. , Ye, Y. , Qiao, Y. , Fu, C.-W. , 2023. ActFloor-GAN: activity-guided adversarial networks for humancentric floorplan design. IEEE Trans. Visual. Comput. Graph. 29 (3), 1610- 1624.

[49] Wang, T.-C. , Liu, M.-Y. , Zhu, J.-Y. , Tao, A. , Kautz, J. , Catanzaro, B. , 2018. High-resolution image synthesis and semantic manipulation with conditional GANs. In: Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Salt Lake City, UT, USA, pp. 8798-8807.

[50] Wu, W. , Fan, L. , Liu, L. , Wonka, P. , 2018. MIQP-based layout design for building interiors. Comput. Graph. Forum 37 (2), 511- 521.

[51] 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), 234: 1- 234: 12.

[52] Xiao, C. , Zhang, M. , Liu, H. , Qin, B. , Huang, B. , 2022. Analysis of the metaverse's spatial restructuring-all databases. Geogr. Geo-Inf. Sci. 38 (2), 1- 9.

[53] Yang, X. , Zhu, C. , 2024. Industrial expert systems review: a comprehensive analysis of typical applications. IEEE Access 12, 88558- 88584.

[54] Yu, Q. , Malaeb, J. , Ma, W. , 2020. Architectural facade recognition and generation through generative adversarial networks. In: Proceedings of 2020 International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE), pp. 310-316.

[55] Zheng, H. , An, K. , Wei, J. , Ren, Y. , 2020. Apartment floor plans generation via generative adversarial networks. In: Proceedings of the 25th Conference for Computer-Aided Architectural Design Research in Asia Conference (CAADRIA), Vol. 2, pp. 599-608. Hong Kong, China.

[56] Zhu, J.Y. , Park, T. , Isola, P. , Efros, A.A. , 2017. Unpaired Image-toImage translation using cycle-consistent adversarial networks.In: 2017 IEEE International Conference on Computer Vision(ICCV), pp. 2242-2251. Venice, Italy.

RIGHTS & PERMISSIONS

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