What type of study is this?

September 10, 2025BuildingsOpen Access

Generative AI for Architectural Façade Design: Measuring Perceptual Alignment Across Geographical, Objective, and Affective Descriptors

View Full Paper

Ask AI

Bookmark

Authors

SLStephen LawUniversity College Lillebaelt CVCleo ValentineHarvard University YKYuval KahlonTokyo Institute of Technology

Discussion

Member takes

Implication

Observational analysis measures perceptual alignment in architectural imagery across geographical and affective descriptors, suggesting bias in AI outputs.

Key Points

Images generated from geographical prompts show strong perceptual alignment with human perceptions, but also reveal biases.
Synthetic images aligned more closely with objective descriptors, such as angular and curvy, than with affective descriptors like utopian or dystopian.
Applying a latent diffusion model for façade generation highlights opportunities and limits in current generative AI models.
Future research should explore cultural biases and enhance understanding of AI's embodied experience with architecture.

Cite This Study

Law et al. (2025) studied this question.

synapsesocial.com/papers/68c18f409b7b07f3a0615f5c https://doi.org/10.3390/buildings15173212

View Full Paper

Ask AI

Bookmark

Implication

Observational analysis measures perceptual alignment in architectural imagery across geographical and affective descriptors, suggesting bias in AI outputs.

Generative AI for Architectural Façade Design: Measuring Perceptual Alignment Across Geographical, Objective, and Affective Descriptors

Discussion

Member takes

Implication

Key Points

Cite This Study

Generative AI for Architectural Façade Design: Measuring Perceptual Alignment Across Geographical, Objective, and Affective Descriptors

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider