Architectural facade design with style and structural features using stable diffusion model

With advancements in digital technology, the field of architectural design has increasingly embraced data and algorithms to enhance design efficiency and quality. Recent advancements in text-to-image (T2I) generation models have enabled the creation of images that correspond to textual descriptions....

Full description

Saved in:
Bibliographic Details
Main Authors: Minghao Wen, Dong Liang, Haibo Ye, Huawei Tu
Format: Article
Language:English
Published: Tsinghua University Press 2024-12-01
Series:Journal of Intelligent Construction
Subjects:
Online Access:https://www.sciopen.com/article/10.26599/JIC.2024.9180034
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846171310232698880
author Minghao Wen
Dong Liang
Haibo Ye
Huawei Tu
author_facet Minghao Wen
Dong Liang
Haibo Ye
Huawei Tu
author_sort Minghao Wen
collection DOAJ
description With advancements in digital technology, the field of architectural design has increasingly embraced data and algorithms to enhance design efficiency and quality. Recent advancements in text-to-image (T2I) generation models have enabled the creation of images that correspond to textual descriptions. However, textual descriptions struggle to capture essential style characteristics in style images. In this study, we proposed a method for architectural facade design based on the stable diffusion model (SDM) that combined stylistic images or keywords as input with the structural conditions of content images to generate images with both stylistic and architectural features. By employing the constrastive language-image pre-training (CLIP) image encoder to convert the style image into its initial image embedding and feature extraction from multilayer cross-attention and training optimization to obtain a pretrained image embedding, the proposed method extracts stylistic features from style images and converts them into corresponding embeddings. This process enables the generated images to embody stylistic features and artistic semantic information. Furthermore, the T2I adapter model is employed to use the architectural structure of content images as conditional guidance, thereby ensuring that the generated images exhibit the corresponding structural features. By leveraging these two aspects, the proposed method can decorate architecture with stylistic features from stylistic images while preserving the architectural structure features of content images, resulting in images that reflect the content images after style transformation. Our method is mainly used in architectural design applications. It was capable of generating facade images from flat design drawings, three-dimensional (3D) architectural models, and hand-drawn sketches and has achieved commendable results.
format Article
id doaj-art-0d1c443eb0724c6ea01cbe4b3c9f6cfa
institution Kabale University
issn 2958-3861
2958-2652
language English
publishDate 2024-12-01
publisher Tsinghua University Press
record_format Article
series Journal of Intelligent Construction
spelling doaj-art-0d1c443eb0724c6ea01cbe4b3c9f6cfa2024-11-11T03:37:19ZengTsinghua University PressJournal of Intelligent Construction2958-38612958-26522024-12-0124918003410.26599/JIC.2024.9180034Architectural facade design with style and structural features using stable diffusion modelMinghao Wen0Dong Liang1Haibo Ye2Huawei Tu3College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaCollege of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaCollege of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaDepartment of Computer Science and Information Technology, La Trobe University, Melbourne 3086, AustraliaWith advancements in digital technology, the field of architectural design has increasingly embraced data and algorithms to enhance design efficiency and quality. Recent advancements in text-to-image (T2I) generation models have enabled the creation of images that correspond to textual descriptions. However, textual descriptions struggle to capture essential style characteristics in style images. In this study, we proposed a method for architectural facade design based on the stable diffusion model (SDM) that combined stylistic images or keywords as input with the structural conditions of content images to generate images with both stylistic and architectural features. By employing the constrastive language-image pre-training (CLIP) image encoder to convert the style image into its initial image embedding and feature extraction from multilayer cross-attention and training optimization to obtain a pretrained image embedding, the proposed method extracts stylistic features from style images and converts them into corresponding embeddings. This process enables the generated images to embody stylistic features and artistic semantic information. Furthermore, the T2I adapter model is employed to use the architectural structure of content images as conditional guidance, thereby ensuring that the generated images exhibit the corresponding structural features. By leveraging these two aspects, the proposed method can decorate architecture with stylistic features from stylistic images while preserving the architectural structure features of content images, resulting in images that reflect the content images after style transformation. Our method is mainly used in architectural design applications. It was capable of generating facade images from flat design drawings, three-dimensional (3D) architectural models, and hand-drawn sketches and has achieved commendable results.https://www.sciopen.com/article/10.26599/JIC.2024.9180034architectural designdigital facade generationstyle transferstable diffusiont2i-adapter
spellingShingle Minghao Wen
Dong Liang
Haibo Ye
Huawei Tu
Architectural facade design with style and structural features using stable diffusion model
Journal of Intelligent Construction
architectural design
digital facade generation
style transfer
stable diffusion
t2i-adapter
title Architectural facade design with style and structural features using stable diffusion model
title_full Architectural facade design with style and structural features using stable diffusion model
title_fullStr Architectural facade design with style and structural features using stable diffusion model
title_full_unstemmed Architectural facade design with style and structural features using stable diffusion model
title_short Architectural facade design with style and structural features using stable diffusion model
title_sort architectural facade design with style and structural features using stable diffusion model
topic architectural design
digital facade generation
style transfer
stable diffusion
t2i-adapter
url https://www.sciopen.com/article/10.26599/JIC.2024.9180034
work_keys_str_mv AT minghaowen architecturalfacadedesignwithstyleandstructuralfeaturesusingstablediffusionmodel
AT dongliang architecturalfacadedesignwithstyleandstructuralfeaturesusingstablediffusionmodel
AT haiboye architecturalfacadedesignwithstyleandstructuralfeaturesusingstablediffusionmodel
AT huaweitu architecturalfacadedesignwithstyleandstructuralfeaturesusingstablediffusionmodel