用于忠实超分辨率的不确定性引导潜在扩散模型
Uncertainty-Guided Latent Diffusion Models for Faithful Super Resolution
- National Taiwan University（国立台湾大学）
- NTU AI-CoRE（台大AI核心研究中心）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对单图像超分辨率的感知-失真权衡问题，提出新型UGDiff不确定性引导潜在扩散模型，通过结合重建不确定性与扩散采样器后验方差引导采样，实现更优的感知-失真平衡，性能优于现有同类方法。
AI中文摘要：
感知-失真权衡是单图像超分辨率（SR）面临的基础挑战。尽管基于扩散的SR方法在生成感知上逼真的图像方面表现出色，但实现高保真度仍是其关键局限。近期基于扩散的SR的进展显示出改善保真度的潜力，但这些方法因高度依赖高保真图像，往往会牺牲感知质量。为解决该问题，我们提出UGDiff，一种旨在进一步改善感知-失真平衡的新型扩散引导范式。具体而言，我们首先估计对应于高保真图像的潜在特征的重建不确定性，随后利用该不确定性引导扩散过程，在高不确定性区域选择性恢复高频细节，同时在其他区域保持保真度。此外，我们的引导方法不仅考虑估计的不确定性，还结合扩散采样器在每个时间步的后验方差，来自适应识别高不确定性区域，这降低了采样后期阶段对高保真图像的依赖，从而实现更好的感知-失真平衡。大量实验结果表明，我们的方法相较于最先进的基于扩散的SR方法表现更优。
英文摘要：
The perception-distortion trade-off poses a fundamental challenge in single-image super-resolution (SR). Although diffusion-based SR methods excel at generating perceptually realistic images, achieving high fidelity remains a key limitation. Recent advances in diffusion-based SR have shown promise in improving fidelity, but these methods often compromise perceptual quality due to their high reliance on a high-fidelity image. To address this, we introduce UGDiff, a novel diffusion guidance paradigm designed to further improve the perception-distortion balance. In particular, we first estimate the reconstruction uncertainty of the latent features corresponding to a high-fidelity image. This uncertainty is then used to guide the diffusion process to selectively restore high-frequency details in high-uncertainty regions, while preserving fidelity elsewhere. Furthermore, our guidance method adaptively identifies the high-uncertainty regions by considering not only the estimated uncertainty but also the posterior variance of the diffusion sampler at each timestep. This relaxes the reliance on the high-fidelity image in the later stages of sampling, thereby achieving a better perception-distortion balance. Extensive experimental results demonstrate that our method performs favorably against state-of-the-art diffusion-based SR methods.
