Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Quality assessment of residential layout designs generated by relational Generative Adversarial Networks (GANs)
Abstract Architectural design, a complex optimization process, employs computational design tools to converge to a subset of design options within a large design space. Recent efforts incorporate deep learning to replace traditional hard-coded rules, yet the nascent and opaque nature of data-driven architectural design requires evaluating its alignment with implicit architectural requirements. This paper assesses whether current data-driven image generation models comply with these non-explicit, domain-specific requirements. Over 80,000 architecturally created floor plans were compared to a similar number of floor plans generated by the authors using the state-of-the-art generative design models. This assessment was based on metrics such as the distribution of space sizes based on their shared/private nature, visibility based on privacy requirements, and room connectivities. The employment of Mann-Whitney U tests and descriptive statistics revealed significant disparities, indicating a gap in the generative design models' understanding of the nuanced, latent rules present in real-world data. These findings highlight the need to develop domain-specific metrics to evaluate the true performance of generative design models, offering insights that will aid in the adoption and advancement of data-driven algorithms in the architecture design domain.
Highlights Verification on deep learning-based floor plan generation model. Focusing on two rules of thumb in floor plan design: 1) the shared nature of spaces is proportional, and 2) the level of privacy is determined by space allocation. Discrepancies in generated vs. real floorplans reveal the algorithm's partial learning of real data rules. Supporting the need for domain-specific performance metrics to better evaluate the algorithm's capacity in design.
Quality assessment of residential layout designs generated by relational Generative Adversarial Networks (GANs)
Abstract Architectural design, a complex optimization process, employs computational design tools to converge to a subset of design options within a large design space. Recent efforts incorporate deep learning to replace traditional hard-coded rules, yet the nascent and opaque nature of data-driven architectural design requires evaluating its alignment with implicit architectural requirements. This paper assesses whether current data-driven image generation models comply with these non-explicit, domain-specific requirements. Over 80,000 architecturally created floor plans were compared to a similar number of floor plans generated by the authors using the state-of-the-art generative design models. This assessment was based on metrics such as the distribution of space sizes based on their shared/private nature, visibility based on privacy requirements, and room connectivities. The employment of Mann-Whitney U tests and descriptive statistics revealed significant disparities, indicating a gap in the generative design models' understanding of the nuanced, latent rules present in real-world data. These findings highlight the need to develop domain-specific metrics to evaluate the true performance of generative design models, offering insights that will aid in the adoption and advancement of data-driven algorithms in the architecture design domain.
Highlights Verification on deep learning-based floor plan generation model. Focusing on two rules of thumb in floor plan design: 1) the shared nature of spaces is proportional, and 2) the level of privacy is determined by space allocation. Discrepancies in generated vs. real floorplans reveal the algorithm's partial learning of real data rules. Supporting the need for domain-specific performance metrics to better evaluate the algorithm's capacity in design.
Quality assessment of residential layout designs generated by relational Generative Adversarial Networks (GANs)
Park, Keundeok (Autor:in) / Ergan, Semiha (Autor:in) / Feng, Chen (Autor:in)
06.12.2023
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Applications of Generative Adversarial Networks (GANs): An Updated Review
| Online Contents | 2019
| DOAJ | 2022