English(EN) Di$^2$CycleSB: Towards High-Quality Unsupervised Nighttime Visibility Enhancement via Schr\"odinger Bridge Transformer

新型变换器框架无监督增强夜间可见性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员推出了一种新颖的无监督夜间可见性增强框架Di$^2$CycleSB。该方法利用由动态积分图像先验引导的Cycle Schrödinger Bridge Transformer来解决光照效应污染带来的挑战。该框架包含一个新的光照效应估计器，该估计器适应类高斯先验，以及一个利用变换器块内这些表示的先验知情生成器。实验表明，Di$^2$CycleSB在没有正则化或分解的情况下有效抑制了光照效应，在真实世界的数据集上实现了视觉上令人愉悦的增强。 AI

影响引入了一种新颖的无监督夜间图像增强方法，有望改善低光照条件下的计算机视觉应用。

排序理由这是一篇详细介绍图像增强新模型和框架的研究论文。[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: 24 / 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增强新模型和框架的研究论文。[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, model release; 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) · Hanting Li, Xin Sun, Wei Ye, Jungong Han, Liang-jie Zhang · 2026-09-01 04:00

Di$^2$CycleSB：通过Schr\"odinger Bridge Transformer实现高质量无监督夜间可见性增强

arXiv:2608.29043v1 Announce Type: new Abstract: Light-effect contamination poses a significant challenge to nighttime visibility enhancement. Most methods suppress light effects by estimating and decomposing them through prior-driven regularization, yet they are often limited by …

本文如何被排名

报道来源 [1]

Di$^2$CycleSB：通过Schr\"odinger Bridge Transformer实现高质量无监督夜间可见性增强

相关实体

相关话题