UniCSG：通过分阶段语义和频率解耦实现统一的高保真内容约束风格驱动生成
UniCSG: Unified High-Fidelity Content-Constrained Style-Driven Generation via Staged Semantic and Frequency Disentanglement
AI总结：
UniCSG通过分阶段语义和频率解耦，解决风格迁移中内容与风格纠缠问题，提升生成内容的忠实度和稳定性。
摘要：
风格迁移必须在保持内容语义的同时匹配目标风格。基于DiT的扩散模型常因内容-风格纠缠导致参考内容泄露和生成不稳定。我们提出了UniCSG，一个统一的框架，用于文本引导和参考引导下的内容约束、风格驱动生成。UniCSG采用分阶段训练：(i) 一个潜在空间语义解耦阶段，结合低频预处理与条件损坏，鼓励内容-风格分离；(ii) 一个潜在空间频率感知细节重建阶段，通过多尺度频率监督细化细节。我们进一步结合像素空间奖励学习，使潜在目标与感知质量对齐。实验表明，在两种设置中均实现了改进的内容忠实度、风格对齐性和鲁棒性。
英文摘要：
Style transfer must match a target style while preserving content semantics. DiT-based diffusion models often suffer from content-style entanglement, leading to reference-content leakage and unstable generation. We present UniCSG, a unified framework for content-constrained, style-driven generation in both text-guided and reference-guided settings. UniCSG employs staged training: (i) a latent-space semantic disentanglement stage that combines low-frequency preprocessing with conditioning corruption to encourage content-style separation, and (ii) a latent-space frequency-aware detail reconstruction stage that refines details via multi-scale frequency supervision. We further incorporate pixel-space reward learning to align latent objectives with perceptual quality after decoding. Experiments demonstrate improved content faithfulness, style alignment, and robustness in both settings.
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。
