风格专家混合模型用于多样化图像风格化
Mixture of Style Experts for Diverse Image Stylization
AI总结：
本文提出StyleExpert框架，基于MoE模型实现语义感知的图像风格化，通过统一编码器和动态门控机制处理多层级语义风格，实验表明其在保持语义和材质细节方面优于现有方法。
摘要：
基于扩散模型的风格化技术已取得显著进展，但现有方法局限于颜色驱动的转换，忽视了复杂的语义和材质细节。我们引入StyleExpert，一种基于混合专家（MoE）的语义感知框架。我们的框架采用统一的风格编码器，训练于大规模内容-风格-风格化三元组数据集，将多样化风格嵌入一致的潜在空间。此嵌入用于条件化一个相似性感知的门控机制，动态路由风格至MoE架构中的专用专家。利用此MoE架构，我们的方法能够处理跨越多个语义层级的多样化风格，从浅层纹理到深层语义。广泛实验表明，StyleExpert在保持语义和材质细节方面优于现有方法，同时能泛化到未见过的风格。我们的代码和收集的图像可在项目页面获取：https://hh-lg.github.io/StyleExpert-Page/。
英文摘要：
Diffusion-based stylization has advanced significantly, yet existing methods are limited to color-driven transformations, neglecting complex semantics and material details. We introduce StyleExpert, a semantic-aware framework based on the Mixture of Experts (MoE). Our framework employs a unified style encoder, trained on our large-scale dataset of content-style-stylized triplets, to embed diverse styles into a consistent latent space. This embedding is then used to condition a similarity-aware gating mechanism, which dynamically routes styles to specialized experts within the MoE architecture. Leveraging this MoE architecture, our method adeptly handles diverse styles spanning multiple semantic levels, from shallow textures to deep semantics. Extensive experiments show that StyleExpert outperforms existing approaches in preserving semantics and material details, while generalizing to unseen styles. Our code and collected images are available at the project page: https://hh-lg.github.io/StyleExpert-Page/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
