提升真实世界图像超分质量，细节更清晰、画质更真实。
FiDeSR: High-Fidelity and Detail-Preserving One-Step Diffusion Super-Resolution
- 通过自适应加权策略聚焦易错区域，训练时增强细节学习。
- 推理阶段引入频域自适应增强，无需重训练即可灵活优化。
- 采用残差噪声修正机制，显著提升细微结构恢复能力。
基于扩散模型的方法在真实世界图像超分辨率（SR）中取得了显著进展。然而，现有方法仍难以同时保持精细细节和高保真重建，常导致视觉质量不佳。本文提出FiDeSR，一种高保真且细节保留的一步式扩散超分辨率框架。训练阶段引入细粒度感知加权策略，自适应强化模型预测误差较高的区域；推理阶段，低频与高频自适应增强器进一步优化重建结果，无需模型重训练即可实现灵活控制；为进一步提升重建精度，FiDeSR采用残差嵌套噪声精修机制，修正扩散过程中的噪声预测误差，增强微小细节恢复能力。相较于现有扩散基方法，FiDeSR在真实世界超分任务上表现更优，生成结果兼具高感知质量与内容忠实性。源代码将发布于：https://github.com/Ar0Kim/FiDeSR。
原文摘要 · Abstract (English)
Diffusion-based approaches have recently driven remarkable progress in real-world image super-resolution (SR). However, existing methods still struggle to simultaneously preserve fine details and ensure high-fidelity reconstruction, often resulting in suboptimal visual quality. In this paper, we propose FiDeSR, a high-fidelity and detail-preserving one-step diffusion super-resolution framework. During training, we introduce a detail-aware weighting strategy that adaptively emphasizes regions where the model exhibits higher prediction errors. During inference, low- and high-frequency adaptive enhancers further refine the reconstruction without requiring model retraining, enabling flexible enhancement control. To further improve the reconstruction accuracy, FiDeSR incorporates a residual-in-residual noise refinement, which corrects prediction errors in the diffusion noise and enhances fine detail recovery. FiDeSR achieves superior real-world SR performance compared to existing diffusion-based methods, producing outputs with both high perceptual quality and faithful content restoration. The source code will be released at: https://github.com/Ar0Kim/FiDeSR.
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。