Three-dimensional (3D) model reconstruction is used in an increasing number of fields related to construction, such as urban planning, mobile communication planning, and solar power assessment. Existing 3D reconstruction models mostly focus on precise measurements, such as laser scanning and ultrasonic mapping. Although these methods can achieve very precise results, they require specific equipment, which is typically expensive. The essence of 3D reconstruction is to infer the overall view of a building through pictures from previously taken perspectives, thereby obtaining pictures from unfamiliar perspectives. In this study, the rendering method is adopted as the starting point, and architectural features are learned by training a neural network to provide the necessary information for rendering. Unlike the more popular projection-based raster rendering method, this study uses a point-based volume rendering method and light sampling to detect architectural features. This rendering method requires the color and density of specific sampling points. Therefore, this study attempts to train a neural network to fit a five-dimensional (5D) function. The input to this function is a 5D vector, including the position (x, y, z) and viewing direction (θ, φ), and the output is the color and density of this point when viewed from this direction. This study adopts the positional encoding method, which reduces the scale of the network and increases both the training and rendering speeds. Our method can train a usable network in dozens of seconds and render a building at 30–60 frames per second.

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.