用最优传输理论提升无配对去雾效果，生成更清晰图像。
When Schrödinger Bridge Meets Real-World Image Dehazing with Unpaired Training
- 基于薛定谔桥建立雾霾与清晰图像的分布映射，减少迭代步数。
- 在多个真实场景数据集上优于现有方法，结构细节保留更佳。
- 结合CLIP模型实现雾霾感知的视觉语言对齐，适合图像增强研究者。
近期基于GAN的无配对去雾方法在处理真实世界雾霾图像方面表现出色。然而，由于生成器运输映射能力有限，难以充分发挥无配对训练的优势。为此，我们提出基于薛定谔桥的新型无配对去雾框架DehazeSB。通过最优传输（OT）理论，直接建立雾霾图像与清晰图像之间的分布桥梁，实现从雾霾到清晰图像的高效映射，显著减少所需步骤并生成高质量结果。为保证重建图像的结构信息和细节一致性，引入细节保持正则化，强制输入雾霾图与输出去雾图在像素级对齐。此外，提出一种新型提示学习策略，利用预训练的CLIP模型实现雾霾感知的视觉-语言对齐。在多个真实世界数据集上的大量实验表明，该方法性能领先。代码已开源：https://github.com/ywxjm/DehazeSB。
原文摘要 · Abstract (English)
Recent advancements in unpaired dehazing, particularly those using GANs, show promising performance in processing real-world hazy images. However, these methods tend to face limitations due to the generator's limited transport mapping capability, which hinders the full exploitation of their effectiveness in unpaired training paradigms. To address these challenges, we propose DehazeSB, a novel unpaired dehazing framework based on the Schrödinger Bridge. By leveraging optimal transport (OT) theory, DehazeSB directly bridges the distributions between hazy and clear images. This enables optimal transport mappings from hazy to clear images in fewer steps, thereby generating high-quality results. To ensure the consistency of structural information and details in the restored images, we introduce detail-preserving regularization, which enforces pixel-level alignment between hazy inputs and dehazed outputs. Furthermore, we propose a novel prompt learning to leverage pre-trained CLIP models in distinguishing hazy images and clear ones, by learning a haze-aware vision-language alignment. Extensive experiments on multiple real-world datasets demonstrate our method's superiority. Code: https://github.com/ywxjm/DehazeSB.
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。