Research on AIGC empowering digital cultural and creative design style transfer and diversified generation methods

• Published in: Applied mathematics and nonlinear sciences Vol. 10; no. 1
• Main Author: Jia, Ran
• Format: Journal Article
• Language: English
• Published: Beirut Sciendo 01.01.2025; De Gruyter Poland
• Subjects: 68T01; AIGC; Artificial intelligence; Digital cultural creation; Improved CycleGAN; Style migration
• ISSN: 2444-8656; 2444-8656
• DOI: 10.2478/amns-2025-0791

The development of artificial intelligence brings new opportunities and challenges to the field of artistic creation. This paper fully analyzes the characteristics and advantages of AIGC-enabled cultural and creative design, and selects CycleGAN algorithm among AI algorithms to carry out cultural and creative design style migration processing. The original CycleGAN algorithm has been improved to construct the AIGC cultural and creative style migration model based on the improved CycleGAN. Take the Forbidden City cultural creations as an example to carry out style migration experiments, and explore the effect of this paper’s improved CycleGAN model on cultural creation style migration and image generation from the objective evaluation and subjective evaluation of the improved CycleGAN-based AIGC cultural creation style migration model. In the experiments of converting original images into cartoon style, oil painting style and new Chinese style, the PSNR value, MS-SSIM value and Per-pixel acc value of this paper’s improved CycleGAN model are all the largest among all the comparison models, and the MSE value is the smallest among all the models, which achieves the optimal objective evaluation results. In subjective evaluation, the improved CycleGAN model in this paper has a comprehensive score of 89.22, which is excellent in style migration and image generation for cultural and creative products.