This paper suggests the potential application of generative artificial intelligence-based image generation technology in the field of architecture, for early phase shape planning, using the styles of renowned architects. The study employed the following approaches: 1) Intensive image generation based on the styles of 20 architects to test the AI's recognition ability and image quality. 2) Additional training was conducted for architects with low recognition rates to construct an enhanced learning model in the quality of image generation. 3) In addition to generating architectural visualization images using existing architects' design styles, alternative styles were proposed through design combinations, aiming to concretize ambiguous idea communication in the early stages of design and enhance its efficiency. The study sheds light on the future prospects of applying this generative AI model in the field of architecture
Yonsei University, Korea (the Republic of) - ORCID: 0009-0002-5362-328X
Yonsei University, Korea (the Republic of)
Yonsei University, Korea (the Republic of) - ORCID: 0000-0003-2009-0376
Korea Advanced Institute of Science Technology, Korea (the Republic of) - ORCID: 0009-0004-7001-2346
Yonsei University, Korea (the Republic of) - ORCID: 0000-0002-5179-6550
Titolo del capitolo
Generative Design Intuition from the Fine-Tuned Models of Named Architects’ Style
Autori
Youngjin Yoo, Hyun Jeong, Youngchae Kim, SeungHyun Cha, Jin-Kook Lee
DOI
10.36253/979-12-215-0289-3.91
Opera sottoposta a peer review
Anno di pubblicazione
2023
Copyright
© 2023 Author(s)
Licenza d'uso
Licenza dei metadati
Titolo del libro
CONVR 2023 - Proceedings of the 23rd International Conference on Construction Applications of Virtual Reality
Sottotitolo del libro
Managing the Digital Transformation of Construction Industry
Curatori
Pietro Capone, Vito Getuli, Farzad Pour Rahimian, Nashwan Dawood, Alessandro Bruttini, Tommaso Sorbi
Opera sottoposta a peer review
Anno di pubblicazione
2023
Copyright
© 2023 Author(s)
Licenza d'uso
Licenza dei metadati
Editore
Firenze University Press
DOI
10.36253/979-12-215-0289-3
eISBN (pdf)
979-12-215-0289-3
eISBN (xml)
979-12-215-0257-2
Collana
Proceedings e report
ISSN della collana
2704-601X
e-ISSN della collana
2704-5846