Towards bridging shape and graph grammars through topology: Fundamental shape processes for architectural representations in artificial intelligence

Dounas, Theodoros and Jabi, Wassim ORCID: https://orcid.org/0000-0002-2594-9568 2025. Towards bridging shape and graph grammars through topology: Fundamental shape processes for architectural representations in artificial intelligence. Presented at: eCAADe 2025, Ankara, Turkiye, 1-5 September 2025. Published in: Sorguç, Arzu Gönenç, Yemişcioğlu, Müge Kruşa, Erol, Serda Buket, Bük, Mustafa Eren, Güney, Dilara, Sulayıc, Aktaş and Akol, Mert eds. eCAADe 2025 Confluence, Volume 1: Proceedings of the 43rd Conference on Education and Research in Computer Aided Architectu ral Design in Europe. pp. 663-672.

Full text not available from this repository.

Abstract

The emergence of transformer models has revitalised AI research in architectural design, despite 3D generative AI lagging behind its 2D counterparts due to dataset limitations. Lightweight frameworks like Topologic enable rapid creation of 3D datasets suitable for graph machine learning, leveraging the library's ability to produce vector‑based geometric, topologic, and semantic models. Our paper presents a topology‑driven graph grammar that generates architecturally meaningful 3D models to support the development of generative AI training datasets. By bridging 3D shapes, topology, and graphs, our system applies shape‑transformation rules procedurally to create 3D models from simple graphs. This approach integrates shape grammars with topology, maintaining semantics while generating novel solutions suitable for 3D generative AI.

Item Type:	Conference or Workshop Item - published (Paper)
Date Type:	Publication
Status:	Published
Schools:	Schools > Architecture
ISBN:	9789491207396
ISSN:	2684-1843
Last Modified:	12 Mar 2026 10:45
URI:	https://orca.cardiff.ac.uk/id/eprint/183697

Actions (repository staff only)

Edit Item