Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 22, 2025Open Access

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

View Full Paper
Ask AI
Bookmark
Share

Authors

SLStephen LawUniversity College LillebaeltCVCleo ValentineHarvard UniversityYKYuval KahlonTokyo Institute of Technology

Discussion

Loading...

Member takes

Implication

Analysis of synthetic architectural façades reveals alignment with geographical, objective, and affective descriptors, indicating underlying biases.

Key Points

  • Images generated with geolocation prompts show notable alignment, yet reveal regional biases in design perception.
  • Results indicate that façades based on objective descriptors align more closely with human perceptions than those from affective descriptors.
  • Employing a latent-diffusion-model, both AI-based and human evaluations assessed the perceptual alignment of generated images.
  • Opportunities and limitations of generative AI in architecture highlight potential data biases and gaps in user experience.

Cite This Study

Law et al. (2025) studied this question.

synapsesocial.com/papers/68af5f19ad7bf08b1eae2375https://doi.org/10.20944/preprints202508.1669.v1
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Generative AI for Architectural Façade Design: Measuring Perceptual Alignment Across Geographical, Objective, and Affective Descriptors2025 · 5 citations
  2. 2Generative AI in architecture: Examining text-to-image models and platforms for cultural heritage representation in UAE design2026
  3. 3Generative artificial intelligence and building design: early photorealistic render visualization of façades using local identity-trained models2024 · 68 citations
  4. 4Generative AI-powered architectural exterior conceptual design based on the design intent2024 · 44 citations
  5. 5Preference and decision difficulty in AI-generated architectural facade stimuli: evidence from spatial design students in Taiwan2026