AI 中文总结
研究基于扩散的真实世界图像超分辨率方法的局限，提出难度感知动态路由（DDR）策略，设计难度估计器并构建不同容量网络，实验证明该模型相比现有方法在效率和有效性上更具优势。
AI 中文摘要
基于扩散的方法通过利用大型预训练稳定扩散（SD）模型作为强大的生成先验，在真实世界图像超分辨率（Real-ISR）中取得了令人瞩目的性能。然而，这些方法仍面临两个关键限制。一是现有基于SD的一步和多步Real-ISR方法对所有输入样本采用统一处理范式，忽略了图像间不同的恢复难度。二是SD模型中VAE的激进分辨率降低（如8倍下采样）导致精细尺度细节不可逆转的损失，后续扩散过程无法恢复。为解决这些限制，我们提出了一种难度感知动态路由（DDR）策略，克服了僵化的一刀切处理范式。具体而言，我们首先设计一个难度估计器来预测每个输入图像的恢复成本，实现自动分配到合适容量的网络。然后，通过调整SD主干中VAE的空间下采样率，构建一组具有不同模型容量的Real-ISR网络，从而在保持简单输入效率的同时，为具有挑战性的情况保留更多高频信息。大量实验表明，与最近的先进方法相比，所提出模型具有更高的效率和有效性。
英文摘要
Diffusion-based methods have achieved impressive performance in real-world image super-resolution (Real-ISR) by leveraging large pre-trained stable diffusion (SD) models as powerful generative priors. However, these methods still face two key limitations. First, existing SD-based one-step and multi-step Real-ISR approaches adopt a unified processing paradigm for all input samples, ignoring the varying restoration difficulty across images. Second, the aggressive resolution reduction of the VAE in SD models (e.g., 8x downsampling) leads to irreversible loss of fine-scale details, which cannot be recovered by the subsequent diffusion process. To address these limitations, we propose a Difficulty-aware Dynamic Routing (DDR) strategy that overcomes the rigid, one-size-fits-all processing paradigm. Specifically, we first design a difficulty estimator to predict the restoration cost of each input image, enabling automatic assignment to a network of appropriate capacity. Then, we construct a set of Real-ISR networks with varying model capacities by modulating the spatial downsampling ratio of the VAE in the SD backbone, thereby preserving more high-frequency information for challenging cases while maintaining efficiency for simpler inputs. Extensive experiments have demonstrated the superior efficiency and effectiveness of the proposed model compared to recent state-of-the-art methods.
CommentsICML 2026 under review
