研究人员开发了TUDSR，一个利用两阶段扩散过程实现更高分辨率的图像超分辨率新框架。该方法解决了当前扩散模型在极端上采样比率和高分辨率输出方面存在的局限性。TUDSR基于SD2.1-base构建，在高达$2048^2$的分辨率下生成高质量图像，表现出最先进的性能。此外，另一篇论文介绍了一种用于图像超分辨率的蒸馏方法RSD，该方法实现了单步恢复，并以更少的计算资源获得了具有竞争力的感知质量。 AI
影响 这些在用于超分辨率的扩散模型方面的进展可能带来更高效、更高质量的图像生成工具。
排序理由 该集群包含两篇研究论文，详细介绍了使用扩散模型进行图像超分辨率的新方法。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
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-resoluti…
arXiv:2503.13358v5 Announce Type: replace Abstract: Diffusion models for super-resolution (SR) produce high-quality visual results but require expensive computational costs. Despite the development of several methods to accelerate diffusion-based SR models, some (e.g., SinSR) fai…
arXiv:2606.09608v1 Announce Type: new Abstract: 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 resolu…
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-resoluti…