AI 中文总结
本研究针对现有扩散模型人脸修复的结构漂移与保真度问题，提出无训练框架WaveFreqAnchor，通过ASWG、MWFI、SHE设计实现高质量人脸修复，性能优于现有方法。
AI 中文摘要
基于扩散模型的人脸修复通过调整预训练扩散模型的采样轨迹已取得显著进展，但现有方法在反向扩散过程中约束不足，导致严重降质下出现与身份相关的结构漂移和保真度下降。为解决该问题，我们提出WaveFreqAnchor，这是一种基于波结构锚定和频率校正扩散的无训练框架。具体而言，锚定空间波结构引导（ASWG）通过各向异性波响应一致性约束人脸结构；多尺度小波-傅里叶注入（MWFI）通过替换预测低频子带的相位，使其与观测结果对齐，校正反向扩散过程中累积的不一致性。针对真实场景，我们进一步引入子带高频增强（SHE），对预测的高频子带执行有界的空间掩码细化，以在未知复合降质下恢复人脸细节。这些设计共同有效保留人脸身份，同时恢复清晰逼真的人脸细节。大量实验表明，我们的方法始终优于现有方法，实现了高质量、高保真的人脸修复。
英文摘要
Diffusion-based face restoration that adjusts the sampling trajectory of pre-trained diffusion models has achieved remarkable progress. However, existing approaches provide insufficient constraints during reverse diffusion, causing identity-related structural drift and degraded fidelity under severe degradations. To address this, we propose WaveFreqAnchor, a training-free framework based on Wave-Structural Anchoring and Frequency Correction Diffusion. Specifically, Anchor-Space Wave-Structural Guidance (ASWG) constrains facial structures through anisotropic wave-response consistency, while Multi-scale Wavelet-Fourier Injection (MWFI) aligns the predicted low-frequency subband with the observation by replacing its phase, correcting inconsistencies accumulated during reverse diffusion. For real-world scenes, we further introduce Subband High-Frequency Enhancement (SHE), which performs bounded, spatially masked refinement on the predicted high-frequency subbands to recover fine facial details under unknown compound degradations. Together, these designs effectively preserve facial identity while restoring sharp and realistic facial details. Extensive experiments show that our method consistently outperforms existing methods, achieving high-quality and high-fidelity face restoration.
Commentstraining-free wavelet-structural diffusion sampling framework for face restoration that improves structural stability, identity consistency, and real-world perceptual quality without task-specific fine-tuning
