PhoenixSR：生成式异构蒸馏释放高效模型用于真实世界超分辨率
PhoenixSR: Generative Heterogeneous Distillation Unleashes Efficient Models for Real-World Super-Resolution
查看机构详情
- University of Science and Technology of China（中国科学技术大学）
- Huawei Technologies Ltd.（华为技术有限公司）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
提出PhoenixSR生成式异构蒸馏框架，将扩散先验通过分布匹配迁移至前馈SR网络，训练后移除扩散组件，在多个基准和骨干网络上实现感知质量提升且保持保真度。
中文摘要 AI 辅助
真实世界图像超分辨率（SR）需要从复杂的低分辨率观测中恢复感知上真实的高分辨率图像，同时保持内容忠实。基于扩散的SR方法受益于强大的生成先验，但会产生大量计算开销，而前馈CNN和Transformer SR模型虽然高效，却往往难以恢复真实的高频细节。这引出了一个自然的问题：能否在不引入扩散组件的情况下，将扩散先验迁移到现有的无扩散SR网络中？为此，我们提出了PhoenixSR，一种生成式异构蒸馏框架，通过基于分数的分布匹配将扩散先验迁移到独立设计的前馈SR网络中。PhoenixSR不是对齐异构特征或模仿采样的扩散输出，而是使用预训练的扩散模型作为分布级监督，同时配对SR监督保持重建保真度。为了使分布匹配对保真度敏感的SR有效，我们引入了异构分布适应，将目标分数适应到SR域，改进对演变中的学生分布的跟踪，并用配对监督锚定训练。我们进一步采用方向可靠性加权，一种基于残差一致性的轻量级重加权策略，以减少不稳定的分布引导。训练后所有扩散相关组件被移除，保留原始学生架构和推理成本不变。在三个SR基准和六个前馈骨干网络（包括SwinIR、HAT、Real-ESRGAN和SeeMoRe）上的实验显示，在基本保持重建保真度的同时，感知质量持续提升。
英文摘要
Real-world image super-resolution (SR) requires recovering perceptually realistic high-resolution images from complex low-resolution observations while preserving faithful content. Diffusion-based SR benefits from strong generative priors but incurs substantial computational overhead, whereas feed-forward CNN and Transformer SR models are efficient yet often struggle to recover realistic high-frequency details. This motivates a natural question: can diffusion priors be transferred to existing diffusion-free SR networks without introducing diffusion components at inference time? To this end, we propose PhoenixSR, a generative heterogeneous distillation framework that transfers diffusion priors to independently designed feed-forward SR networks through score-based distribution matching. Rather than aligning heterogeneous features or imitating sampled diffusion outputs, PhoenixSR uses the pretrained diffusion model as distribution-level supervision, while paired SR supervision preserves reconstruction fidelity. To make distribution matching effective for fidelity-sensitive SR, we introduce Heterogeneous Distribution Adaptation, which adapts the target score to the SR domain, improves tracking of the evolving student distribution, and anchors training with paired supervision. We further employ Directional Reliability Weighting, a lightweight residual-consistency-based reweighting strategy that reduces unstable distributional guidance. All diffusion-related components are removed after training, leaving the original student architecture and inference cost unchanged. Experiments on three SR benchmarks and six feed-forward backbones, including SwinIR, HAT, Real-ESRGAN, and SeeMoRe, show consistent perceptual improvements with largely preserved reconstruction fidelity.
