arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.07255cs.CVeess.IV
收藏

面向真实世界超分辨率的真实数据生成研究

Towards Realistic Data Generation for Real-World Super-Resolution

USTC(中国科学技术大学);HKUST(GZ)(香港科技大学(广州));CUHK(香港中文大学)

Long Peng, Wenbo Li, Renjing Pei, Jingjing Ren, Jiaqi Xu, Yang Wang, Yang Cao, Zheng-Jun Zha

AI总结:

本文提出无监督数据生成框架RealDGen,通过内容-退化解耦扩散模型从不成对真实图像中生成大规模高质量配对数据,显著提升真实世界超分辨率模型性能。

摘要:

现有的图像超分辨率(SR)技术在复杂真实场景中往往难以有效泛化,原因在于训练数据与实际场景之间存在显著差异。为应对这一挑战,以往工作要么手动模拟复杂的基于物理的退化过程,要么采用基于学习的技术,但这些方法仍难以同时生成大规模、真实且多样化的数据。本文提出一种新颖的真实解耦数据生成器(RealDGen),这是一个面向真实世界超分辨率的无监督学习数据生成框架。我们精心设计了内容提取与退化提取策略,并将其整合到一个新颖的内容-退化解耦扩散模型中,从而利用不成对的真实低分辨率(LR)和高分辨率(HR)图像生成真实的低分辨率图像。大量实验表明,RealDGen 在生成大规模、高质量且能反映真实世界退化的配对数据方面表现出色,显著提升了主流 SR 模型在多个真实世界基准上的性能。

英文摘要:

Existing image super-resolution (SR) techniques often fail to generalize effectively in complex real-world settings due to the significant divergence between training data and practical scenarios. To address this challenge, previous efforts have either manually simulated intricate physical-based degradations or utilized learning-based techniques, yet these approaches remain inadequate for producing large-scale, realistic, and diverse data simultaneously. In this paper, we introduce a novel Realistic Decoupled Data Generator (RealDGen), an unsupervised learning data generation framework designed for real-world super-resolution. We meticulously develop content and degradation extraction strategies, which are integrated into a novel content-degradation decoupled diffusion model to create realistic low-resolution images from unpaired real LR and HR images. Extensive experiments demonstrate that RealDGen excels in generating large-scale, high-quality paired data that mirrors real-world degradations, significantly advancing the performance of popular SR models on various real-world benchmarks.

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。

↑