Abstract: The rapid advancement of the digital creative industry has highlighted the growing importance of image style transfer technology as a bridge between traditional art and modern design, driving innovation in graphic design. However, conventional style transfer methods face significant challenges, including low computational efficiency and unnatural style transformation in complex image scenarios. This study addresses these limitations by introducing a novel approach to image style transfer based on the MLP-Mixer model. Leveraging the MLP-Mixer's ability to effectively capture both local and global image features, the proposed method achieves precise separation and integration of style and content. Experimental results demonstrate that the MLP-Mixer-based style transfer significantly enhances the naturalness and diversity of style transformation while preserving image clarity and detail. Additionally, the processing speed is improved by 50%, with style conversion accuracy and user satisfaction increasing by 30% and 35%, respectively, compared to traditional methods. These findings underscore the potential of the MLP-Mixer model for advancing efficiency and realism in graphic design applications.

Qibin Wang, Xiao Chen and Huan Su, “Application of MLP-Mixer-Based Image Style Transfer Technology in Graphic Design” International Journal of Advanced Computer Science and Applications(IJACSA), 16(1), 2025. http://dx.doi.org/10.14569/IJACSA.2025.0160159

@article{Wang2025,
title = {Application of MLP-Mixer-Based Image Style Transfer Technology in Graphic Design},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2025.0160159},
url = {http://dx.doi.org/10.14569/IJACSA.2025.0160159},
year = {2025},
publisher = {The Science and Information Organization},
volume = {16},
number = {1},
author = {Qibin Wang and Xiao Chen and Huan Su}
}