.png){kind=icon}
Return
Architectural layout generation using a graph-constrained conditional Generative Adversarial Network (GAN)
.png){kind=icon}
PRE
AI
DOI：10.1016/j.autcon.2023.105053.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Efficiently generating appealing and realistic architectural space configurations has been a significant challenge for designers. This paper presents a deep-learning approach, providing architects with increased control over the final design outcomes. Employing deep learning algorithms to analyze the graph structure of input bubble diagrams facilitates the generation of node-based space layouts confined within predefined borders, ensuring a balance between creative freedom and practical constraints. The findings reveal the effectiveness of the graph-constrained data-driven method in automating the space layout design process. Automating space arrangement accelerates the building design workflow, yielding more efficient and productive results for architects.
Keywords:
Generative design
Space layout
Graph network
Deep learning
GAN
Journal
IF:
11.5
Papers:
6.3K
Citations:
4.2W
Organization
No organization information available
Cited Papers
Self-focusing effects of dispositional self-consciousness, mirror presence, and audience presence.
.png){kind=icon}
PREAI
ShareSaveShareSaveAnalysis of Maximum-Likelihood Sequence Estimation Performance for Quadrature Amplitude Modulation
PREAI
ShareSave
ShareSaveShareSaveShareSaveShareSaveShareSave
View more