Abstract
Real-image super-resolution (Real-ISR) seeks to recover HR images from LR inputs with mixed, unknown degradations. While diffusion models surpass GANs in perceptual quality, they under-reconstruct high-frequency (HF) details due to a low-frequency (LF) bias and a depth-wise "low-first, high-later" hierarchy. We introduce FRAMER, a plug-and-play training scheme that exploits diffusion priors without changing the backbone or inference. At each denoising step, the final-layer feature map teaches all intermediate layers. Teacher and student feature maps are decomposed into LF/HF bands via FFT masks to align supervision with the model's internal frequency hierarchy. For LF, an Intra Contrastive Loss (IntraCL) stabilizes globally shared structure. For HF, an Inter Contrastive Loss (InterCL) sharpens instance-specific details using random-layer and in-batch negatives. Two adaptive modulators, Frequency-based Adaptive Weight (FAW) and Frequency-based Alignment Modulation (FAM), reweight per-layer LF/HF signals and gate distillation by current similarity. Across U-Net and DiT backbones (e.g., Stable Diffusion 2, 3), FRAMER consistently improves PSNR/SSIM and perceptual metrics (LPIPS, NIQE, MANIQA, MUSIQ). Ablations validate the final-layer teacher and random-layer negatives.
Abstract (translated)
真实图像超分辨率（Real-ISR）旨在从含有混合未知退化的低分辨率输入中恢复高分辨率图像。虽然扩散模型在感知质量上超越了生成对抗网络（GANs），但由于它们对低频信号的偏好和深度方向上的“先低后高”层次结构，这些模型难以准确重构高频细节。我们引入了一种名为FRAMER的即插即用训练方案，该方案利用了扩散模型的先验知识，而不改变主干网络或推断过程。 在每个去噪步骤中，最终层特征图会指导所有中间层。教师和学生特征图通过FFT掩码被分解成低频（LF）/高频（HF）带，以确保监督与模型内部频率层次结构对齐。对于低频信号，我们引入了一个内对比损失（Intra Contrastive Loss, IntraCL），它有助于稳定全局共享的结构；而对于高频信号，则使用随机层和批次内的负样本来增强特定实例的细节，即交互式对比损失（Inter Contrastive Loss, InterCL）。 此外，还引入了两种自适应调制器：基于频率的自适应权重（Frequency-based Adaptive Weight, FAW）和基于频率对齐调节（Frequency-based Alignment Modulation, FAM）。FAW重新调整每个层次中的低频/高频信号的权重，而FAM通过当前相似度来控制知识蒸馏的过程。 在U-Net和DiT主干网络（如Stable Diffusion 2、3）上进行实验时，FRAMER方案在峰值信噪比（PSNR）、结构相似性指数（SSIM），以及感知质量指标（LPIPS、NIQE、MANIQA、MUSIQ）等多个方面都得到了一致的改进。消融研究验证了最终层教师和随机层负样本的有效性。 该方法通过利用扩散模型的优势并克服其固有的局限性，为图像超分辨率领域提供了一种新的训练策略，具有广泛的应用前景。
URL
https://arxiv.org/abs/2512.01390
