Urban space simulation based on wave function collapse and convolutional neural network

Lin, Bo, Jabi, Wassim ORCID: https://orcid.org/0000-0002-2594-9568 and Diao, Rongdan 2020. Urban space simulation based on wave function collapse and convolutional neural network. Presented at: Symposium on Simulation for Architecture and Urban Design (SimAUD 2020), Online, 25-28 May 2020. Proceedings of the 11th Annual Symposium on Simulation for Architecture and Urban Design. (Articl) Society for Computer Simulation International, pp. 1-8. 10.5555/3465085.3465103

Preview

PDF - Published Version Download (1MB) | Preview

Abstract

In this paper, we propose a pipeline of urban space synthesis which leverages Wave Function Collapse (WFC) and Convolutional Neural Networks (CNNs) to train the computer how to design urban space. Firstly, we establish an urban design database. Then, the urban road networks, urban block spatial forms and urban building function layouts are generated by WFC and CNNs and evaluated by designer afterwards. Finally, the 3D models are generated. We demonstrate the feasibility of our pipeline through the case study of the North Extension of Central Green Axis in Wenzhou. This pipeline improves the efficiency of urban design and provides new ways of thinking for architecture and urban design.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Publication
Status:	Published
Schools:	Architecture
Publisher:	Society for Computer Simulation International
Date of First Compliant Deposit:	4 March 2022
Last Modified:	15 May 2023 10:13
URI:	https://orca.cardiff.ac.uk/id/eprint/148029

Actions (repository staff only)

Edit Item