Generating Daylight-driven Architectural Design via Diffusion Models

Pengzhi Li, Baijuan Li
4/20/2024
cs.CV

No official code found. Best implementation: unlimitedli/DDADesign (22 stars, MIT, last commit 2025-02-09).

Abstract

In recent years, the rapid development of large-scale models has made new possibilities for interdisciplinary fields such as architecture. In this paper, we present a novel daylight-driven AI-aided architectural design method. Firstly, we formulate a method for generating massing models, producing architectural massing models using random parameters quickly. Subsequently, we integrate a daylight-driven facade design strategy, accurately determining window layouts and applying them to the massing models. Finally, we seamlessly combine a large-scale language model with a text-to-image model, enhancing the efficiency of generating visual architectural design renderings. Experimental results demonstrate that our approach supports architects' creative inspirations and pioneers novel avenues for architectural design development. Project page: https://zrealli.github.io/DDADesign/.

View on arXivView PDF

Code Implementations(1)

unlimitedli/DDADesignCommunity74%
MIT

Generating Daylight-driven Architectural Design via Diffusion Models

222Apr 20, 20241 years agoMIT

Similar Papers

Sketch-to-Architecture: Generative AI-aided Architectural Design

81% match2024

CoMa: Contextual Massing Generation with Vision-Language Models

80% match2026

Tokenization Allows Multimodal Large Language Models to Understand, Generate and Edit Architectural Floor Plans

78% match2026

Exploring Gen-AI applications in building research and industry: A review

78% match2024

HouseTune: Two-Stage Floorplan Generation with LLM Assistance

78% match2024

Encoding Urban Ecologies: Automated Building Archetype Generation through Self-Supervised Learning for Energy Modeling

78% match2024

Cite this paper

@article{li2024generating,
  title  = {Generating Daylight-driven Architectural Design via Diffusion Models},
  author = {Pengzhi Li and Baijuan Li},
  year   = {2024},
  eprint = {2404.13353},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2404.13353v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2404.13353.svg)](https://codeofpaper.com/paper/2404.13353)

Discussion