StyleGallery：面向任意图像参考的免训练且语义感知的个性化风格迁移
StyleGallery: Training-free and Semantic-aware Personalized Style Transfer from Arbitrary Image References
用潜扩散特征做自适应聚类得到语义区域，再以块过滤匹配和区域风格损失引导扩散采样，为免训练风格迁移提供了可复用的区域级对齐与能量函数优化范式。
原文依据 semantic region segmentation (adaptive clustering on latent diffusion features to divide regions without extra inputs)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论尽管基于扩散模型的图像风格迁移取得了进展，现有方法通常受限于：1）语义鸿沟：风格参考可能缺失恰当的内容语义，导致风格化不可控；2）依赖额外约束（如语义掩码），限制了适用性；3）刚性的特征关联缺乏自适应的全局—局部对齐，无法在细粒度风格化与全局内容保持之间取得平衡。这些局限，尤其是无法灵活利用风格输入，从根本上限制了风格迁移在个性化、准确性和适应性方面的表现。为解决这些问题，我们提出 StyleGallery，一个免训练且语义感知的框架，支持任意参考图像作为输入并实现有效的个性化定制。它包含三个核心阶段：语义区域分割（在潜扩散特征上做自适应聚类以划分区域，无需额外输入）；聚类区域匹配（对提取的特征做块过滤以实现精确对齐）；以及风格迁移优化（以能量函数引导的扩散采样，配合区域风格损失来优化风格化）。在我们提出的基准上的实验表明，StyleGallery 在内容结构保持、区域风格化、可解释性和个性化定制方面优于最优方法，尤其是在利用多个风格参考时。源代码与数据集见 https://github.com/iiiiiiiword/StyleGallery。
英文原文摘要 1535 字符
Despite the advancements in diffusion-based image style transfer, existing methods are commonly limited by 1) semantic gap: the style reference could miss proper content semantics, causing uncontrollable stylization; 2) reliance on extra constraints (e.g., semantic masks) restricting applicability; 3) rigid feature associations lacking adaptive global-local alignment, failing to balance fine-grained stylization and global content preservation. These limitations, particularly the inability to flexibly leverage style inputs, fundamentally restrict style transfer in terms of personalization, accuracy, and adaptability.To address these, we propose StyleGallery, a training-free and semantic-aware framework that supports arbitrary reference images as input and enables effective personalized customization.It comprises three core stages: semantic region segmentation (adaptive clustering on latent diffusion features to divide regions without extra inputs); clustered region matching (block filtering on extracted features for precise alignment); and style transfer optimization (energy function-guided diffusion sampling with regional style loss to optimize stylization).Experiments on our introduced benchmark demonstrate that StyleGallery outperforms state-of-the-art methods in content structure preservation, regional stylization, interpretability, and personalized customization, particularly when leveraging multiple style references. The source code and dataset are available at https://github.com/iiiiiiiword/StyleGallery.