English(EN) Beyond MMSE: Enhancing PnP Restoration with ProxiMAP

ProxiMAP通过改进MAP近似来增强成像恢复

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-09 05:00

研究人员推出了一种新颖的迭代近似方法ProxiMAP，旨在增强成像逆问题中的即插即用（PnP）恢复技术。该新方法旨在通过更好地针对最大后验（MAP）估计来解决MMSE去噪器常见的过度平滑问题。ProxiMAP在当前迭代的残差噪声水平下查询去噪器，使去噪器保持在分布内，并在去模糊和超分辨率等各种应用中提高感知质量，有时会以牺牲PSNR为代价。为了提高效率，还提出了一种单步蒸馏和混合调度，以降低计算成本的方式保留了显著的改进。 AI

影响 ProxiMAP在图像恢复任务中提供了改进的感知质量，可能推动AI驱动的成像应用的能力。

排序理由这是一篇详细介绍图像恢复新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇详细介绍图像恢复新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Kenta Vert, Giacomo Meanti, Scott Pesme, Michael Arbel, Julien Mairal · 2026-11-09 05:00

超越MMSE：使用ProxiMAP增强PnP恢复

arXiv:2605.16396v2 Announce Type: replace Abstract: Plug-and-Play (PnP) methods for imaging inverse problems replace the proximal operator of the prior, a maximum a posteriori (MAP) denoiser, with a pretrained minimum mean squared error (MMSE) denoiser. MMSE estimates are oversmo…

本文如何被排名

报道来源 [1]

超越MMSE：使用ProxiMAP增强PnP恢复

相关实体

相关话题