A Training-Free Style-Personalization via SVD-Based Feature Decomposition
Lee, Kyoungmin, Park, Jihun, Gim, Jongmin, Choi, Wonhyeok, Hwang, Kyumin, Kim, Jaeyeul, Im, Sunghoon
原文摘要
We present a training-free framework for style-personalized image generation that operates during inference using a scale-wise autoregressive model. Our method generates a stylized image guided by a single reference style while preserving semantic consistency and mitigating content leakage. Through a detailed step-wise analysis of the generation process, we identify a pivotal step where the dominant singular values of the internal feature encode style-related components. Building upon this insight, we introduce two lightweight control modules: Principal Feature Blending, which enables precise modulation of style through SVD-based feature reconstruction, and Structural Attention Correction, which stabilizes structural consistency by leveraging content-guided attention correction across fine stages. Without any additional training, extensive experiments demonstrate that our method achieves competitive style fidelity and prompt fidelity compared to fine-tuned baselines, while offering faster inference and greater deployment flexibility.
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表