发表机构
Indian Institute of Science（印度科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究基于预训练深度去噪器的图像重建不稳定问题，提出数据驱动稳定框架，通过局部量形式化不稳定性并自适应正则化，引入可训练收缩算子控制不稳定IR算子，实验证明该方法提升了PnP和RED重建的性能与可靠性。
AI 中文摘要
预训练的深度去噪器可用于通过即插即用（PnP）和去噪正则化（RED）算法解决各种基于模型的图像重建任务，无需针对每个任务重新训练。这些去噪器仅针对单步去噪进行训练。在迭代过程中将它们用作图像重建（IR）正则化器会使重建不稳定。一种常见的失败模式是峰值和崩溃行为：诸如PSNR之类的指标在早期迭代中有所改善，然后突然下降，这使得这些算法在实践中不可靠。我们提出了一个数据驱动的稳定框架，该框架通过一个局部量形式化任何IR算子的这种不稳定性，并通过自适应地正则化这个量来防止崩溃，无需重新训练或修改给定的预训练网络。我们的关键思想是用一个收缩算子控制潜在不稳定的IR算子，其稳定迭代充当锚并防止崩溃。我们进一步引入了一个有效的可训练收缩算子家族，它们在保持轻量级的同时充当强大的锚。在近端算法、去噪器架构、噪声水平和成像任务上的广泛实验表明，PnP和RED重建具有一致的、无崩溃的性能和更高的可靠性。
英文摘要
Pretrained deep denoisers can be used to solve a wide range of model-based image reconstruction tasks via Plug-and-Play (PnP) and Regularization-by-Denoising (RED) algorithms, without retraining per task. These denoisers are trained only for single-step denoising. Using them as Image Reconstruction (IR) regularizers in an iterative process can destabilize reconstruction. A common failure mode is the peak-and-collapse behaviour: metrics such as PSNR improve for early iterations and then abruptly degrade, making these algorithms unreliable in practice. We propose a data-driven stabilization framework that (i) formalizes this instability of any IR operator through a local quantity and (ii) prevents collapse by regularizing this quantity adaptively, requiring no retraining or modification of the given pretrained network. Our key idea is to control the potentially unstable IR operator with a contractive operator whose stable iterates act as an anchor and prevent collapse. We further introduce an efficient family of trainable contractive operators that serve as strong anchors while remaining lightweight. Extensive experiments across proximal algorithms, denoiser architectures, noise levels, and imaging tasks show consistent, collapse-free performance and improved reliability of PnP and RED reconstruction.
CommentsAccepted at ECCV 2026
