為噪聲著色：基於對抗性索博列夫對齊的忠實影像超解析度

摘要

生成先驗在圖像超解析度（SR）中常損害忠實還原，我們將此限制歸因於各向同性目標與自然圖像流形之間的根本性光譜失配。儘管直接偏好優化提供了對齊途徑，但其依賴於光譜平坦的高斯雜訊，無法區分真實高頻細節與幻覺。為填補此幾何鴻溝，我們提出ASASR，一個具理論基礎的框架，通過明確對雜訊轉移核進行著色以鏡像自然光譜衰減，從而將生成流重新塑造成Sobolev誘導的黎曼幾何。為驅動此幾何對齊，我們整合了一個基於Riesz表示定理的參數化對抗機制，該機制合成相當於最壞情況Sobolev梯度的目標負樣本，從而沿可行結構失敗的切空間引導優化。廣泛評估顯示，ASASR在保持光譜一致性和結構忠實度上優於領先的生成基線，提供了一個有效減輕偽影的穩健解決方案。

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.