通过动态调度风格注入位置与强度，显著提升图像风格迁移的品质与保真度。
Scheduled Style Injection: Expanding the Style-Content Pareto Frontier in Training-Free Diffusion-based Style Transfer
- 在扩散模型不同层和去噪阶段动态调整风格注入强度，突破固定权衡限制。
- 最佳配置实现ArtFID 27.036，比StyleID提升6.1%，全范围表现更优。
- 无需训练、零参数，仅需几行代码即可部署，适合快速实验与应用。
基于预训练扩散模型的风格迁移进展迅速，但一个核心问题仍被忽视：应在模型何处最强地注入风格？当前主流的训练自由方法StyleID采用单一全局参数（gamma）在所有层和时间步均匀作用，导致风格质量与内容保留之间的固定权衡。我们系统探索了四个控制维度：在解码器层间、去噪时间步间，以及沿两个轴调度ControlNet几何条件。结果一致显示：递减调度（浅层和早期时间步更强结构信号注入）优于反向。此外，调度形状也关键：余弦与平方根调度优于线性。最重要的是，gamma调度与ControlNet条件几乎独立。结合配置后扩展了帕累托前沿，相比任意单一下调设置均表现更优。最佳平衡配置达到ArtFID 27.036，较StyleID的28.801提升6.1%，且在整个风格-内容权衡前沿保持一致优势。35种配置共生成超28,000张风格化图像，使用四种互补指标验证。结果在多种SD骨干网络上具有一致排序。所有改进均为训练自由、参数自由，仅需少量调度代码；代码已开源于https://github.com/ameyskulkarni/scheduled_style_injection。
原文摘要 · Abstract (English)
Style transfer with pre-trained diffusion models has advanced rapidly, but a core question remains underexplored: where in the model should style injection be strongest? StyleID, the leading training-free method, uses a single global parameter (gamma) uniformly across all layers and timesteps, which forces a fixed tradeoff between style quality and content preservation. We show this tradeoff is unnecessarily rigid. We systematically explore four dimensions of control: varying style injection strength across decoder layers, across denoising timesteps, and scheduling ControlNet geometric conditioning along both axes. The pattern is consistent everywhere: decreasing schedules, with stronger structural signal injection in shallower layers and earlier timesteps, reliably outperform the reverse. Beyond direction, schedule shape matters: cosine and square-root timestep schedules outperform linear. Most importantly, we find that gamma scheduling and ControlNet conditioning are nearly independent. The resulting combined configurations expand the Pareto frontier, offering superior tradeoffs between style fidelity and content preservation compared to any single baseline setting. Our best balanced configuration achieves ArtFID of 27.036 versus StyleID's 28.801 - a 6.1% relative improvement, with consistent gains across the full style-content tradeoff frontier. Results are validated across 35 configurations totaling over 28,000 stylized images using four complementary metrics. These findings generalize across SD backbones with identical rank ordering. All modifications are training-free, parameter-free, and require only a few lines of scheduling code; code is available at https://github.com/ameyskulkarni/scheduled_style_injection.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。