用几何对齐方法让修复模型适应新场景，无需配对数据
Generative Manifold Distillation: Aligning Restoration Trajectories with Natural Image Prior
- 通过流匹配将模糊图像投影到自然图像流形上生成伪真值
- 仅需低质量输入即可显著提升感知质量，且无额外计算开销
- 适合缺乏标注数据的现实图像修复任务，尤其适用于部署场景
预训练的图像修复模型在分布外的真实世界退化上表现不佳。由于真实数据缺乏成对的高质量-低质量样本，且无监督方法常需不稳定的结构改动，域适应面临挑战。本文提出生成流形蒸馏（GMD），将域适应重新定义为几何流形对齐问题。GMD 在严格无配对设置下运行，仅需目标域的低质量（LQ）观测。借助冻结的文生图基础模型的流匹配动态，GMD 将偏离流形的修复结果投影回自然图像流形，生成高质量伪目标。为确保稳定性，引入质量门控流形过滤器，剔除偏离流形的样本；同时采用源锚定轨迹正则化，防止误差累积。最终，GMD 将强大的生成先验蒸馏至高效修复网络。实验表明，仅使用 LQ 输入，GMD 即可无缝适配新分布，显著提升感知质量，且无需架构修改或增加推理延迟。
原文摘要 · Abstract (English)
Pre-trained image restoration models often fail on out-of-distribution (OOD) real-world degradations. Adapting to these domains is challenging as real-world data lacks paired ground truth, and unsupervised methods often require unstable architectural changes. We propose Generative Manifold Distillation (GMD), which reframes domain adaptation as geometric manifold alignment. GMD operates in a strictly unpaired setting, requiring only low-quality (LQ) target observations. By leveraging the flow-matching dynamics of a frozen text-to-image foundation model, GMD projects off-manifold restorations onto the natural image manifold to generate high-quality pseudo-targets. To ensure stability, a quality-gated manifold filter rejects off-manifold samples, while source-anchored trajectory regularization prevents error accumulation. Ultimately, GMD distills a powerful generative prior into an efficient restoration network. Experiments demonstrate that GMD seamlessly adapts to new distributions using only LQ inputs, drastically improving perceptual quality with zero architectural modifications or added inference latency.
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。