Balanced Image Stylization with Style Matching Score
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916854852747264 |
|---|---|
| author | Jiang, Yuxin Jiang, Liming Yang, Shuai Liu, Jia-Wei Tsang, Ivor Shou, Mike Zheng |
| author_facet | Jiang, Yuxin Jiang, Liming Yang, Shuai Liu, Jia-Wei Tsang, Ivor Shou, Mike Zheng |
| contents | We present Style Matching Score (SMS), a novel optimization method for image stylization with diffusion models. Balancing effective style transfer with content preservation is a long-standing challenge. Unlike existing efforts, our method reframes image stylization as a style distribution matching problem. The target style distribution is estimated from off-the-shelf style-dependent LoRAs via carefully designed score functions. To preserve content information adaptively, we propose Progressive Spectrum Regularization, which operates in the frequency domain to guide stylization progressively from low-frequency layouts to high-frequency details. In addition, we devise a Semantic-Aware Gradient Refinement technique that leverages relevance maps derived from diffusion semantic priors to selectively stylize semantically important regions. The proposed optimization formulation extends stylization from pixel space to parameter space, readily applicable to lightweight feedforward generators for efficient one-step stylization. SMS effectively balances style alignment and content preservation, outperforming state-of-the-art approaches, verified by extensive experiments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_07601 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Balanced Image Stylization with Style Matching Score Jiang, Yuxin Jiang, Liming Yang, Shuai Liu, Jia-Wei Tsang, Ivor Shou, Mike Zheng Computer Vision and Pattern Recognition Machine Learning We present Style Matching Score (SMS), a novel optimization method for image stylization with diffusion models. Balancing effective style transfer with content preservation is a long-standing challenge. Unlike existing efforts, our method reframes image stylization as a style distribution matching problem. The target style distribution is estimated from off-the-shelf style-dependent LoRAs via carefully designed score functions. To preserve content information adaptively, we propose Progressive Spectrum Regularization, which operates in the frequency domain to guide stylization progressively from low-frequency layouts to high-frequency details. In addition, we devise a Semantic-Aware Gradient Refinement technique that leverages relevance maps derived from diffusion semantic priors to selectively stylize semantically important regions. The proposed optimization formulation extends stylization from pixel space to parameter space, readily applicable to lightweight feedforward generators for efficient one-step stylization. SMS effectively balances style alignment and content preservation, outperforming state-of-the-art approaches, verified by extensive experiments. |
| title | Balanced Image Stylization with Style Matching Score |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2503.07601 |