TUDSR：用于更高超分辨率的两次上采样扩散
TUDSR: Twice Upsampling-Diffusion for Higher Super-Resolution
提出两阶段上采样扩散框架，通过循环分块训练策略在有限资源下实现高分辨率（2048²）图像超分，可复用于其他需要突破模型原生分辨率限制的生成任务。
原文依据 we present TUDSR, a Twice Upsampling-Diffusion framework for higher SR
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于扩散的生成模型在真实世界图像超分辨率（SR）中取得了显著成功。借助分块扩散技术，这些模型能够生成超出其原生支持分辨率的高分辨率图像。然而，此类高分辨率（如2048²）输出的质量往往极差，我们认为主要源于两个因素：图像上采样比例（如x8）超过模型原生支持的上采样比例（如x4），以及模型的原生支持分辨率。实践中，训练原生高分辨率模型需要更大的架构，这会带来显著的计算开销和GPU内存成本，在资源有限的设备上难以实现。因此，我们提出TUDSR，一种用于更高超分辨率的两次上采样扩散框架。TUDSR框架主要包含两个阶段：第一阶段在R分辨率下训练，第二阶段在NR分辨率下引入基于循环分块的训练策略。每个阶段均采用由生成器和判别器组成的一步生成对抗网络架构。基于SD2.1-base，我们开发了TUDSR-S，在多个基准上达到最优性能。大量实验进一步表明，TUDSR-S能够生成1024²甚至2048²分辨率的高质量图像，显著优于现有方法。代码见 https://github.com/wuer5/TUDSR。
英文原文摘要 1414 字符
Diffusion-based generative models have achieved remarkable success in real-world image super-resolution (SR). With tiled diffusion techniques, these models can produce high-resolution images that exceed their native-supported resolution. However, the quality of such high-resolution (e.g 2048^2) outputs often remains extremely poor, primarily due to two factors we consider: the image upsampling ratio (e.g x8) exceeding the model's native-supported upsampling ratio (e.g x4), and the model's native-supported resolution. In practice, training a native high-resolution model requires larger architectures, which incur significant computational overhead and GPU memory costs, making it hard on limited-resource equipment. Thus, we present TUDSR, a Twice Upsampling-Diffusion framework for higher SR. The TUDSR framework mainly consists of two stages: the first involves training at R-resolution, and the second introduces a looped chunk-based training strategy at NR-resolution. Each stage adapts a one-step GAN architecture comprising a generator and a discriminator. Based on SD2.1-base, we develop TUDSR-S, which achieves state-of-the-art performance across multiple benchmarks. Extensive experiments further demonstrate that TUDSR-S generates high-quality images at the resolutions of 1024^2 and even 2048^2, significantly outperforming existing approaches. Code is available at https://github.com/wuer5/TUDSR.