Towards a Robust Evaluation Framework for Generative Urban Design

Haya Brama, Agata Dalach, Tal Grinshpoun, Jonathan Dortheimer

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Thispaper critically reviews the evaluation methods employed in the Generative Urban Design (GUD) literature. The review reveals various evaluation methods, including human-based, performance-based, and statistical evaluation· An analysis of the evaluation methods shows that each approach has limitations, and none fully addresses the unique challenges of evaluating GUD. The paper concludes that more robust and comprehensive evaluation methods are neededfor GUD.

Original languageEnglish
Title of host publicationProceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2024
EditorsOdysseas Kontovourkis, Marios C. Phocas, Gabriel Wurzer
PublisherEducation and research in Computer Aided Architectural Design in Europe
Pages529-538
Number of pages10
ISBN (Print)9789491207372
DOIs
StatePublished - 2024
Event42nd Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2024 - Nicosia, Cyprus
Duration: 9 Sep 202413 Sep 2024

Publication series

NameProceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe
Volume1
ISSN (Print)2684-1843

Conference

Conference42nd Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2024
Country/TerritoryCyprus
CityNicosia
Period9/09/2413/09/24

Keywords

  • DeepLeammg,GAN
  • FID score
  • Generative Urban Design
  • Machine-Learning

Fingerprint

Dive into the research topics of 'Towards a Robust Evaluation Framework for Generative Urban Design'. Together they form a unique fingerprint.

Cite this