A Question of Style

Campo, Matias del; Manninger, Sandra; Carlson, Alexandra

doi:10.1007/978-981-15-6568-7_11

Cited by 6 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We show that even in its current early prototype stage, the system supports design explorations such as object morphing, object addition and rudimentary 3D style transfer through the use of semantic labels. With regard to style and style transfer Del Campo et al 32 note that “features learned by a neural network… can allow us as designers to rethink the low level visual components that comprise a given ‘style’ and how neural representations of style can be used as a method of architectural interrogation” [32, p 183].…”

Section: Discussionmentioning

confidence: 99%

Using deep learning to generate design spaces for architecture

Sebestyen

Hirschberg

Rasoulzadeh

2023

International Journal of Architectural Computing

View full text Add to dashboard Cite

We present an early prototype of a design system that uses Deep Learning methodology—Conditional Variational Autoencoders (CVAE)—to arrive at custom design spaces that can be interactively explored using semantic labels. Our work is closely tied to principles of parametric design. We use parametric models to create the dataset needed to train the neural network, thus tackling the problem of lacking 3D datasets needed for deep learning. We propose that the CVAE functions as a parametric tool in itself: The solution space is larger and more diverse than the combined solution spaces of all parametric models used for training. We showcase multiple methods on how this solution space can be navigated and explored, supporting explorations such as object morphing, object addition, and rudimentary 3D style transfer. As a test case, we implemented some examples of the geometric taxonomy of “Operative Design” by Di Mari and Yoo.

show abstract

Section: Discussionmentioning

confidence: 99%

Using deep learning to generate design spaces for architecture

Sebestyen

Hirschberg

Rasoulzadeh

2023

International Journal of Architectural Computing

View full text Add to dashboard Cite

show abstract

“…Indeed, these reflections extend beyond the simple methodologies of their use to a fundamental reexamination of the imperative to establish relevant criteria for their evaluation. This transformative influence of artificial intelligence in architectural design echoes a broader discourse in the academic community (Datta et al, 2018;Gkoumas et al, 2020: del Campo, Manninger, S., Carlson, A., 2020. The implications of such technological interventions require conscientious assessment of their ethical, pedagogical, and professional ramifications (O'Neil, 2016).…”

Section: Polis Universitymentioning

confidence: 94%

Venturing into the Age of AI: Insights and Perspectives

2023

FORUM A+P

View full text Add to dashboard Cite

ObjectivesThe aims of the Workshop in Multimedia Design were to combine the content of the multimedia with applications connected to product branding. The theoretical aspects were presented, while at the same time, a series of exercises were performed and presented within the class. The participants worked in teams and at the end of every exercise, they were all actively present the projects' outcomes. The rest of the participants and the curators asked questions in order to clarify issues that came up during the presentations. Additionally, a series of digital tools was presented based on real case scenarios and thus the participants were exposed to knowledges and experiences, when using a multimedia toolkit, with an increased number of tools, from different points of view i.e. text, audio, sketching, vector design, image processing, animation, video, artificial intelligence based media creations.

show abstract

“…This technique significantly reduces labor time compared to modeling hundreds of buildings to achieve structural consistency, which is necessary for other approaches 8,21 and excessive formatting to triangulated mesh faces. 11…”

Section: Methodsmentioning

confidence: 99%

“…10 Graph convolutional networks require triangled meshes as closed volumes and, therefore, can barely extend beyond deforming topological shapes of the same degree. 11…”

Section: Implementation Of Gans In Architecturementioning

confidence: 99%

Design across multi-scale datasets by developing a novel approach to 3DGANs

Ennemoser

Mayrhofer-Hufnagl

2023

International Journal of Architectural Computing

View full text Add to dashboard Cite

The development of Generative Adversarial Networks (GANs) has accelerated the research of Artificial Intelligence (AI) in architecture as a generative tool. However, since their initial invention, many versions have been developed that only focus on 2D image datasets for training and images as output. The current state of 3DGAN research has yielded promising results. However, these contributions focus primarily on building mass, extrusion of 2D plans, or the overall shape of objects. In comparison, our newly developed 3DGAN approach, using fully spatial building datasets, demonstrates that unprecedented interconnections across different scales are possible resulting in unconventional spatial configurations. Unlike a traditional design process, based on analyzing only a few precedents (typology) according to the task, by collaborating with the machine we can draw on a significantly wider variety of buildings across multiple typologies. In addition, the dataset was extended beyond the scale of complete buildings and involved building components that define space. Thus, our results achieve a high spatial diversity. A detailed analysis of the results also revealed new hybrid architectural elements illustrating that the machine continued the interconnections of scale since elements were not explicitly part of the dataset, becoming a true design collaborator.

show abstract

A Question of Style

Cited by 6 publications

References 3 publications

Using deep learning to generate design spaces for architecture

Using deep learning to generate design spaces for architecture

Venturing into the Age of AI: Insights and Perspectives

Design across multi-scale datasets by developing a novel approach to 3DGANs

Contact Info

Product

Resources

About