用多任务微调让去噪器更好解决图像逆问题。
Multi Task Denoiser Training for Solving Linear Inverse Problems
- 在求解过程中同步训练去噪器，适配多个逆问题。
- 平均提升1.34 dB PSNR，且减少迭代次数。
- 适合想用统一去噪器处理多种图像恢复任务的研究者。
插件式先验（PnP）和通过去噪正则化（RED）已证明，图像去噪器可有效替代传统正则项，用于超分辨率、去马赛克和图像修复等线性逆问题求解。文献表明，去噪器的残差与图像对数先验梯度相关，从而支持基于梯度的图像生成（如扩散模型）及新逆问题求解方法。本文在Kadkhodaie与Simoncelli的梯度求解框架基础上，提出在迭代求解过程中对去噪器进行微调。通过在整个求解框架内端到端联合训练，并跨多个任务同时优化，得到一个单一且通用的去噪器，专为逆问题设计。实验显示，即使使用简单基线模型，经此方式微调后，在六个不同逆问题上平均实现+1.34 dB的PSNR提升，同时减少所需迭代次数。进一步分析发现，微调后的去噪器优化目标从最小化标准去噪误差（MMSE）转向更贴近理想先验梯度，专门用于引导逆问题重建。
原文摘要 · Abstract (English)
Plug-and-Play Priors (PnP) and Regularisation by Denoising (RED) have established that image denoisers can effectively replace traditional regularisers in linear inverse problem solvers for tasks like super-resolution, demosaicing, and inpainting. It is now well established in the literature that a denoiser's residual links to the gradient of the image log prior (Miyasawa and Tweedie), enabling iterative, gradient ascent-based image generation (e.g., diffusion models), as well as new methods for solving inverse problems. Building on this, we propose enhancing Kadkhodaie and Simoncelli's gradient-based inverse solvers by fine-tuning the denoiser within the iterative solving process itself. Training the denoiser end-to-end across the solver framework and simultaneously across multiple tasks yields a single, versatile denoiser optimised for inverse problems. We demonstrate that even a simple baseline model fine-tuned this way achieves an average PSNR improvement of +1.34 dB across six diverse inverse problems while reducing the required iterations. Furthermore, we analyse the fine-tuned denoiser's properties, finding that its optimisation objective implicitly shifts from minimising standard denoising error (MMSE) towards approximating an ideal prior gradient specifically tailored for guiding inverse recovery.
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。