StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models

Zichong Chen, Shijin Wang, Yang Zhou*
Visual Computing Research Center, CSSE, Shenzhen University
Eurographics 2025
*Corresponding Author
Paper arXiv Code

TL;DR: A finetuning-based approach for style-specific text-to-image generation that ensures robust style consistency and aligns text semantics.

MY ALT TEXT

Abstract

Synthesizing visually impressive images that seamlessly align both text prompts and specific artistic styles remains a significant challenge in Text-to-Image (T2I) diffusion models. This paper introduces StyleBlend, a method designed to learn and apply style representations from a limited set of reference images, enabling content synthesis of both text-aligned and stylistically coherent. Our approach uniquely decomposes style into two components, composition and texture, each learned through different strategies. We then leverage two synthesis branches, each focusing on a corresponding style component, to facilitate effective style blending through shared features without affecting content generation. StyleBlend addresses the common issues of text misalignment and weak style representation that previous methods have struggled with. Extensive qualitative and quantitative comparisons demonstrate the superiority of our approach.

Method

Gallery of StyleBlend Results

Qualitative Comparisons




Quantitative Comparisons

Applications & Variations


BibTeX

@misc{chen2025styleblend,
  title={StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models}, 
  author={Zichong Chen and Shijin Wang and Yang Zhou},
  year={2025},
  eprint={2502.09064},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2502.09064}, 
}