arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.03352cs.CVcs.AIcs.LG
收藏

基于侧信息的推理时搜索用于扩散模型图像重建

Inference-Time Search Using Side Information for Diffusion-Based Image Reconstruction

Department of Electrical and Computer Engineering, Texas A&M University(电气与计算机工程系,德克萨斯A&M大学)

Mahdi Farahbakhsh, Vishnu Teja Kunde, Dileep Kalathil, Krishna Narayanan, Jean-Francois Chamberland

AI总结:

提出一种即插即用、无需训练的推理时搜索框架,将侧信息融入现有扩散模型逆问题求解器,显著提升重建质量。

摘要:

扩散模型已被用作解决逆问题的先验。然而,现有方法通常忽略了能够显著提高重建质量的侧信息,尤其是在严重病态设置中。在这项工作中,我们提出了一种新颖的框架,通过推理时搜索将侧信息以即插即用、无需训练的方式融入现有的基于扩散模型的逆问题求解器。通过在多种逆问题(包括图像修复、超分辨率和几种去模糊任务)以及多种基于扩散模型的逆问题求解器(DPS、DAPS和MPGD)上的大量实验,我们表明,用我们的框架增强每个求解器,其重建质量始终优于相应的原始方法。为了展示我们方法的通用性,我们考虑了多种形式的侧信息,包括参考图像、文本描述和解剖学MRI扫描。代码可在该仓库中获取:https://github.com/mahdi-farahbakhsh/DISS。

英文摘要:

Diffusion models have been used as priors for solving inverse problems. However, existing approaches typically overlook side information that could significantly improve reconstruction quality, especially in severely ill-posed settings. In this work, we propose a novel framework that incorporates side information into existing diffusion-based inverse problem solvers via inference-time search, in a plug-and-play, training-free manner. Through extensive experiments across a range of inverse problems, including inpainting, super-resolution, and several deblurring tasks, and across multiple diffusion-based inverse problem solvers (DPS, DAPS, and MPGD), we show that augmenting each solver with our framework consistently improves the quality of the reconstructions over the corresponding original method. To demonstrate the generality of our approach, we consider diverse forms of side information, including reference images, textual descriptions, and anatomical MRI scans. The code is available at this \href{https://github.com/mahdi-farahbakhsh/DISS}{repository}\footnote{https://github.com/mahdi-farahbakhsh/DISS}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑