根据噪声水平动态暴露测量频率，提升扩散模型在逆问题中的修复效果。
Posterior Continuation with Noise-Conditioned Frequency Exposure for Diffusion Inverse Problems
- 按当前噪声水平选择性暴露测量频率，构建渐进式后验
- 在超分辨率等任务中最高提升5 dB PSNR，优于强基线
- 适合图像修复、去模糊等需要高保真重建的场景
扩散后验采样通过结合预训练扩散先验与测量一致性引导来解决逆问题。然而，在高噪声水平下，全频段引导可能不可靠，此时干净估计会引入得分误差，且高频测量方向难以识别。我们提出，后验引导应根据瞬时扩散噪声水平暴露测量频率。基于此原则，设计了一种后验延续框架，构建一系列中间后验，其似然函数强调当前可靠频段，并逐步恢复全频段一致性。具体实现采用稳定采样器，结合扩散预测器、频段限制似然优化及哈尔域承诺规则——对可靠粗粒度修正进行承诺，而推迟弱可识别细节。在超分辨率、补全和去模糊任务中，方法表现达到竞争性至最先进水平，尤其在FFHQ与ImageNet上的运动去模糊任务中，相比强基线最高提升5 dB PSNR。
原文摘要 · Abstract (English)
Diffusion posterior sampling solves inverse problems by combining a pretrained diffusion prior with measurement-consistency guidance. However, full-band guidance can be unreliable at high noise levels, where clean estimates contain score-induced errors and high-frequency measurement directions are weakly identifiable. We argue that posterior guidance should expose measurement frequencies according to the instantaneous diffusion noise level. Based on this principle, we propose a posterior continuation framework that constructs a family of intermediate posteriors whose likelihood emphasizes currently reliable frequency bands and gradually returns to full-band consistency. We instantiate this framework with a stabilized sampler that combines a diffusion predictor, frequency-limited likelihood refinement, and a Haar-domain commitment rule that commits reliable coarse corrections while deferring weakly identifiable details. Across super-resolution, inpainting, and deblurring, our method achieves competitive-to-state-of-the-art restoration performance, including up to 5 dB PSNR improvement on motion deblurring over strong baselines in evaluations on FFHQ and ImageNet.
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。