Architectural Facade Design with Style and Structural Features using Stable Diffusion Model

Abstract

With the 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 generation models have enabled the creation of images corresponding to textual descriptions. However, textual descriptions struggle to capture essential style characteristics in style images. In this paper, we propose a method for architectural facade design based on the Stable Diffusion Model, which combines stylistic images or stylistic keywords as input with the structural conditions of content images to generate images with both stylistic and architectural features. By employing the CLIP image encoder to convert the style image into its initial image embedding, utilizing feature extraction from multi-layer cross-attention, and training optimization to obtain a pre-trained image embedding, this method extracts stylistic features from style images and converts them into corresponding embeddings. This process enables the generated images to embody both stylistic features and artistic semantic information. Furthermore, the T2I-Adapter model is employed to use the architectural structure of content images as conditional guidance, ensuring that the generated images exhibit the corresponding structural features. By leveraging these two aspects, this 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 primarily applied in architectural design applications. It is capable of generating facade images from flat design drawings, 3D architectural models, and hand-drawn sketches, and has achieved commendable results.

Topics & Concepts