Abstract
Generative deep learning diffusion models have been attracting mainstream attention in the field of 2D image generation. We propose a prototype which brings a diffusion network into the third dimension, with the purpose of generating geometries for conceptual design. We explore the possibilities of generating 3D datasets, using parametric design to overcome the problem of the current lack of available architectural 3D data suitable for training neural networks. Furthermore, we propose a data representation based on volumetric density grids which is applicable to train diffusion networks. Our early prototype demonstrates the viability of the approach and suggests future options to develop deep learning generative 3D tools for architectural design.
Originalsprache | englisch |
---|---|
Titel | eCAADe 2023 - Digital Design Reconsidered |
Untertitel | Proceedings of the International Conference on Education and Research in Computer Aided Architectural Design in Europe |
Redakteure/-innen | Wolfgang Dokonal, Urs Hirschberg, Gabriel Wurzer, Gabriel Wurzer |
Herausgeber (Verlag) | eCAADe (Education and Research in Computer Aided Architectural Design in Europe) |
Seiten | 451-460 |
Seitenumfang | 10 |
Band | 2 |
ISBN (Print) | 9789491207358 |
DOIs | |
Publikationsstatus | Veröffentlicht - 2023 |
Veranstaltung | 41st Conference on Education and Research in Computer Aided Architectural Design in Europe - Digital Design Reconsidered: eCAADe 2023 - Graz University of Technology, Graz, Österreich Dauer: 20 Sept. 2023 → 22 Sept. 2023 https://ecaade2023.tugraz.at/ |
Konferenz
Konferenz | 41st Conference on Education and Research in Computer Aided Architectural Design in Europe - Digital Design Reconsidered |
---|---|
Kurztitel | eCAADe 2023 |
Land/Gebiet | Österreich |
Ort | Graz |
Zeitraum | 20/09/23 → 22/09/23 |
Internetadresse |
ASJC Scopus subject areas
- Architektur
- Ausbildung bzw. Denomination
- Computergrafik und computergestütztes Design