StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models
No Thumbnail Available
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association and John Wiley & Sons Ltd.
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.
Description
CCS Concepts: Computing methodologies → Image processing; Image representations
@article{10.1111:cgf.70034,
journal = {Computer Graphics Forum},
title = {{StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models}},
author = {Chen, Zichong and Wang, Shijin and Zhou, Yang},
year = {2025},
publisher = {The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {10.1111/cgf.70034}
}