Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN

Ruşen Eroğlu; Leman Figen Gül

doi:10.52842/conf.ecaade.2022.2.575

Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN

Ruşen Eroğlu, Leman Figen Gül

Department of Architecture

Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

In recent years, generative models have been rapidly transforming into a broad field of research, and artificial intelligence (AI) works are increasing. Since deep learning technologies such as Generative Adversarial Networks (GANs) providing synthesized new images are becoming more accessible, researchers in the design and related fields are very much interested in adapting GANs into practice. Especially, StyleGAN has a strong capability for image learning, reconstruction simulation, and absorbing the pixel characteristics of images in the input dataset. StyleGAN also produces similar imitation outputs and summarizes all the input data into one "average output". The study aims to reveal the potential of these outputs that can be employed as a visual inspiration aid for designers. This article will discuss the outputs of the experiments, findings, and prospects of StyleGAN. Keywords: Artificial Intelligence, Machine Learning, Generative Adversarial Networks, StyleGAN.

Original language	English
Title of host publication	eCAADe 2022 - Co-creating the Future
Subtitle of host publication	Inclusion in and through Design
Editors	Burak Pak, Gabriel Wurzer, Rudi Stouffs
Publisher	Education and research in Computer Aided Architectural Design in Europe
Pages	575-582
Number of pages	8
ISBN (Print)	9789491207334
DOIs	https://doi.org/10.52842/conf.ecaade.2022.2.575
Publication status	Published - 2022
Event	40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022 - Ghent, Belgium Duration: 13 Sept 2022 → 16 Sept 2022

Publication series

Name	Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe
Volume	2
ISSN (Print)	2684-1843

Conference

Conference	40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022
Country/Territory	Belgium
City	Ghent
Period	13/09/22 → 16/09/22

Bibliographical note

Publisher Copyright:
© 2022, Education and research in Computer Aided Architectural Design in Europe. All rights reserved.

Access to Document

10.52842/conf.ecaade.2022.2.575

Cite this

Eroğlu, R., & Gül, L. F. (2022). Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN. In B. Pak, G. Wurzer, & R. Stouffs (Eds.), eCAADe 2022 - Co-creating the Future: Inclusion in and through Design (pp. 575-582). (Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe; Vol. 2). Education and research in Computer Aided Architectural Design in Europe. https://doi.org/10.52842/conf.ecaade.2022.2.575

Eroğlu, Ruşen ; Gül, Leman Figen. / Architectural Form Explorations through Generative Adversarial Networks : Predicting the potentials of StyleGAN. eCAADe 2022 - Co-creating the Future: Inclusion in and through Design. editor / Burak Pak ; Gabriel Wurzer ; Rudi Stouffs. Education and research in Computer Aided Architectural Design in Europe, 2022. pp. 575-582 (Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe).

@inproceedings{e8ee06674948410693b2017c76e3e908,

title = "Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN",

abstract = "In recent years, generative models have been rapidly transforming into a broad field of research, and artificial intelligence (AI) works are increasing. Since deep learning technologies such as Generative Adversarial Networks (GANs) providing synthesized new images are becoming more accessible, researchers in the design and related fields are very much interested in adapting GANs into practice. Especially, StyleGAN has a strong capability for image learning, reconstruction simulation, and absorbing the pixel characteristics of images in the input dataset. StyleGAN also produces similar imitation outputs and summarizes all the input data into one {"}average output{"}. The study aims to reveal the potential of these outputs that can be employed as a visual inspiration aid for designers. This article will discuss the outputs of the experiments, findings, and prospects of StyleGAN. Keywords: Artificial Intelligence, Machine Learning, Generative Adversarial Networks, StyleGAN.",

author = "Ru{\c s}en Eroğlu and G{\"u}l, {Leman Figen}",

note = "Publisher Copyright: {\textcopyright} 2022, Education and research in Computer Aided Architectural Design in Europe. All rights reserved.; 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022 ; Conference date: 13-09-2022 Through 16-09-2022",

year = "2022",

doi = "10.52842/conf.ecaade.2022.2.575",

language = "English",

isbn = "9789491207334",

series = "Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe",

publisher = "Education and research in Computer Aided Architectural Design in Europe",

pages = "575--582",

editor = "Burak Pak and Gabriel Wurzer and Rudi Stouffs",

booktitle = "eCAADe 2022 - Co-creating the Future",

address = "Belgium",

}

Eroğlu, R & Gül, LF 2022, Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN. in B Pak, G Wurzer & R Stouffs (eds), eCAADe 2022 - Co-creating the Future: Inclusion in and through Design. Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe, vol. 2, Education and research in Computer Aided Architectural Design in Europe, pp. 575-582, 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022, Ghent, Belgium, 13/09/22. https://doi.org/10.52842/conf.ecaade.2022.2.575

Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN. / Eroğlu, Ruşen; Gül, Leman Figen.
eCAADe 2022 - Co-creating the Future: Inclusion in and through Design. ed. / Burak Pak; Gabriel Wurzer; Rudi Stouffs. Education and research in Computer Aided Architectural Design in Europe, 2022. p. 575-582 (Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe; Vol. 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Architectural Form Explorations through Generative Adversarial Networks

T2 - 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022

AU - Eroğlu, Ruşen

AU - Gül, Leman Figen

PY - 2022

Y1 - 2022

N2 - In recent years, generative models have been rapidly transforming into a broad field of research, and artificial intelligence (AI) works are increasing. Since deep learning technologies such as Generative Adversarial Networks (GANs) providing synthesized new images are becoming more accessible, researchers in the design and related fields are very much interested in adapting GANs into practice. Especially, StyleGAN has a strong capability for image learning, reconstruction simulation, and absorbing the pixel characteristics of images in the input dataset. StyleGAN also produces similar imitation outputs and summarizes all the input data into one "average output". The study aims to reveal the potential of these outputs that can be employed as a visual inspiration aid for designers. This article will discuss the outputs of the experiments, findings, and prospects of StyleGAN. Keywords: Artificial Intelligence, Machine Learning, Generative Adversarial Networks, StyleGAN.

AB - In recent years, generative models have been rapidly transforming into a broad field of research, and artificial intelligence (AI) works are increasing. Since deep learning technologies such as Generative Adversarial Networks (GANs) providing synthesized new images are becoming more accessible, researchers in the design and related fields are very much interested in adapting GANs into practice. Especially, StyleGAN has a strong capability for image learning, reconstruction simulation, and absorbing the pixel characteristics of images in the input dataset. StyleGAN also produces similar imitation outputs and summarizes all the input data into one "average output". The study aims to reveal the potential of these outputs that can be employed as a visual inspiration aid for designers. This article will discuss the outputs of the experiments, findings, and prospects of StyleGAN. Keywords: Artificial Intelligence, Machine Learning, Generative Adversarial Networks, StyleGAN.

UR - http://www.scopus.com/inward/record.url?scp=85139189269&partnerID=8YFLogxK

U2 - 10.52842/conf.ecaade.2022.2.575

DO - 10.52842/conf.ecaade.2022.2.575

M3 - Conference contribution

AN - SCOPUS:85139189269

SN - 9789491207334

T3 - Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe

SP - 575

EP - 582

BT - eCAADe 2022 - Co-creating the Future

A2 - Pak, Burak

A2 - Wurzer, Gabriel

A2 - Stouffs, Rudi

PB - Education and research in Computer Aided Architectural Design in Europe

Y2 - 13 September 2022 through 16 September 2022

ER -

Eroğlu R, Gül LF. Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN. In Pak B, Wurzer G, Stouffs R, editors, eCAADe 2022 - Co-creating the Future: Inclusion in and through Design. Education and research in Computer Aided Architectural Design in Europe. 2022. p. 575-582. (Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe). doi: 10.52842/conf.ecaade.2022.2.575

Architectural Form Explorations through Generative Adversarial Networks: Predicting the potentials of StyleGAN

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this