通过彩色噪声对齐自然图像频谱，提升超分辨率真实还原能力
Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution
- 用苏波列夫几何重构生成过程，让噪声匹配自然图像频谱衰减
- 在多个数据集上超越主流基线，尤其在保留高频细节和结构真实度上显著提升
- 适合关注图像超分辨真实性、对抗训练与几何建模的研究者
图像超分辨率中的生成先验常导致还原失真，我们将其归因于各向同性目标与自然图像流形之间的本质频谱错位。尽管直接偏好优化提供对齐路径，但其依赖频谱平坦的高斯噪声，无法区分真实高频细节与幻觉内容。为此，我们提出ASASR，一个理论驱动的框架，通过显式地‘彩色化’噪声转移核，使生成流程映射到由苏波列夫诱导的黎曼几何中，以模仿自然图像的频谱衰减特性。在此几何基础上，引入基于里斯表示定理的参数化对抗器，生成等价于最坏情况苏波列夫梯度的目标负样本，引导优化沿可接受结构失效的切空间进行。大量实验表明，ASASR在多个基准上优于现有生成基线，尤其在保持频谱一致性与结构保真度方面表现优异，有效缓解了伪影问题。
原文摘要 · Abstract (English)
Generative priors in Image Super-Resolution (SR) often compromise faithful restoration, we attribute this limitation to a fundamental spectral misalignment between isotropic objectives and the intrinsic natural image manifold. While Direct Preference Optimization offers a path to alignment, its reliance on spectrally flat Gaussian noise fails to distinguish authentic high-frequency details from hallucinations. To bridge this geometric gap, we propose ASASR, a theoretically grounded framework that recasts the generative flow into a Sobolev-induced Riemannian geometry by explicitly coloring the noise transition kernel to mirror natural spectral decay. Driving this geometric alignment, we integrate a parametric adversary grounded in the Riesz Representation Theorem, which synthesizes targeted negative samples equivalent to worst-case Sobolev gradients to direct optimization along the tangent space of plausible structural failures. Extensive evaluations demonstrate that ASASR outperforms leading generative baselines, particularly in preserving spectral consistency and structural fidelity, offering a robust solution that effectively mitigates artifacts.
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。