A computational framework for aesthetic preferences in architecture using computer vision and artificial neural networks

Página do item simplificado
dc.contributor.authorSardenberg V.
dc.contributor.authorGuatelli I.
dc.contributor.authorBecker M.
dc.date.accessioned2024-10-01T06:12:03Z
dc.date.available2024-10-01T06:12:03Z
dc.date.issued2024
dc.description.abstract© The Author(s) 2024.This paper introduces a computational aesthetics framework utilizing computer vision (CV) and artificial neural networks (ANN) to predict the aesthetic preferences of groups of people for architecture. It relies on part-to-whole theories from aesthetics and cognitive psychology. A survey of a group of people on preferences of images is held to record an average hedonic response (AHR). CV algorithms MSER and SAM recognize parts in images. Birkhoff’s aesthetic measure formula is adapted by employing the number of parts and their connections. These quantities are used as input layers of an ANN, and the AHR is the target output. The ANN evaluates images to output a predicted hedonic response (PHR), which is tested as a criterion in parametric design space navigation and in mapping the latent space of GANs. We conclude that such a framework is a heuristic method for better understanding the design and latent spaces and exploring designs.
dc.identifier.doi10.1177/14780771241279350
dc.identifier.issnNone
dc.identifier.urihttps://dspace.mackenzie.br/handle/10899/39484
dc.relation.ispartofInternational Journal of Architectural Computing
dc.rightsAcesso Restrito
dc.subject.otherlanguageArtificial neural networks
dc.subject.otherlanguagecomputational aesthetics
dc.subject.otherlanguagecomputer vision
dc.subject.otherlanguagedesign space navigation
dc.subject.otherlanguageempirical aesthetics
dc.subject.otherlanguagehedonic response
dc.subject.otherlanguageheuristics
dc.subject.otherlanguagelatent space map
dc.subject.otherlanguagequantitative aesthetics
dc.titleA computational framework for aesthetic preferences in architecture using computer vision and artificial neural networks
dc.typeArtigo
local.scopus.citations0
local.scopus.eid2-s2.0-85203713801
local.scopus.updated2025-04-01
local.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85203713801&origin=inward
Arquivos
Coleções
Artigos de periódico