This study presents a groundbreaking method named Expo-GAN (Exposition-Generative Adversarial Network) for style transfer in exhibition hall design, using a refined version of the Cycle Generative Adversarial Network (CycleGAN). The primary goal is to enhance the transformation of image styles while maintaining visual consistency, an area where current CycleGAN models often fall short. These traditional models typically face difficulties in accurately capturing expansive features as well as the intricate stylistic details necessary for high-quality image transformation. To address these limitations, the research introduces several key modifications to the CycleGAN architecture. Enhancements to the generator involve integrating U-net with SpecTransformer modules. This integration incorporates the use of Fourier transform techniques coupled with multi-head self-attention mechanisms, which collectively improve the generator’s ability to depict both large-scale structural patterns and minute elements meticulously in the generated images. This enhancement allows the generator to achieve a more detailed and coherent fusion of styles, essential for exhibition hall designs where both broad aesthetic strokes and detailed nuances matter significantly. The study also proposes innovative changes to the discriminator by employing dilated convolution and global attention mechanisms. These are derived using the Differentiable Architecture Search (DARTS) Neural Architecture Search framework to expand the receptive field, which is crucial for recognizing comprehensive artistically styled images. By broadening the ability to discern complex artistic features, the model avoids previous pitfalls associated with style inconsistency and missing detailed features. Moreover, the traditional cyde-consistency loss function is replaced with the Learned Perceptual Image Patch Similarity (LPIPS) metric. This shift aims to significantly enhance the perceptual quality of the resultant images by prioritizing human-perceived similarities, which aligns better with user expectations and professional standards in design aesthetics. The experimental phase of this research demonstrates that this novel approach consistently outperforms the conventional CycleGAN across a broad range of datasets. Complementary ablation studies and qualitative assessments underscore its superiority, particularly in maintaining detail fidelity and style continuity. This is critical for creating a visually harmonious exhibition hall design where every detail contributes to the overall aesthetic appeal. The results illustrate that this refined approach effectively bridges the gap between technical capability and artistic necessity, marking a significant advancement in computational design methodologies.

Although existing style transfer techniques have made significant progress in the field of image generation, there are still some challenges in the field of exhibition hall design. The existing style transfer methods mainly focus on the transformation of single dimensional features, but ignore the deep integration of content and style features in exhibition hall design. In addition, existing methods are deficient in detail retention, especially in accurately capturing and reproducing local textures and details while preserving the content image structure. In addition, point-based attention mechanisms tend to ignore the complexity and diversity of image features in multi-dimensional space, resulting in alignment problems between features in different semantic areas, resulting in inconsistent stylistic features in content areas. In this context, this paper proposes a semantic-enhanced multimodal style transfer algorithm tailored for exhibition hall design. The proposed approach leverages a multimodal encoder architecture to integrate information from text, source images, and style images, using separate encoder modules for each modality to capture shallow, deep, and semantic features. A novel Style Transfer Convolution (STConv) convolutional kernel, based on the Visual Geometry Group (VGG) 19 network, is introduced to improve feature extraction in style transfer. Additionally, an enhanced Transformer encoder is incorporated to capture contextual semantic information within images, while the CLIP model is employed for text data processing. A hybrid attention module is designed to precisely capture style features, achieving multimodal feature fusion via a diffusion model that generates exhibition hall design images aligned with stylistic requirements. Quantitative experiments show that compared with the most advanced algorithms, the proposed method has achieved significant performance improvement on both Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) indexes. For example, on the ExpoArchive dataset, the proposed method has a FID value of 87.9 and a KID value of 1.98, which is significantly superior to other methods.