arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.21045cs.CVcs.AI
收藏

LPNSR: 通过可学习噪声预测实现最优噪声引导扩散图像超分辨率

LPNSR: Learnable Noise Prediction for Diffusion-Based Image Super-Resolution

Huazhong University of Science and Technology(华中科技大学)

Shuwei Huang, Shizhuo Liu, Zijun Wei

AI总结:

本文提出LPNSR,通过可学习噪声预测器替代随机高斯噪声,优化扩散模型中的中间噪声,实现高感知性能,无需依赖大规模文本到图像先验。

摘要:

基于扩散的图像超分辨率(SR)旨在从低分辨率(LR)观测中重建高分辨率(HR)图像。然而,反向扩散过程中的内在随机性导致扩散基于SR模型的性能在不同采样运行中变化显著,特别是在将采样轨迹压缩为有限步数时。本文建立了一个理论框架,从最大似然估计的角度推导出扩散模型中最优中间噪声的闭式解析解,揭示了一种一致的条件依赖结构,该结构适用于各种扩散范式。我们在此框架下实例化残差位移扩散范式,并相应设计了一个LR引导的多输入感知噪声预测器。我们进一步通过高质量的预上采样网络缓解初始化偏差。紧凑的4步轨迹唯一地使整个反向链的端到端优化成为可能,这在传统长轨迹扩散模型中计算上是不可行的。大量实验表明,LPNSR在合成和现实数据集上均实现了最先进的感知性能,且不依赖任何大规模文本到图像先验。我们的方法源代码可在https://github.com/Faze-Hsw/LPNSR找到。

英文摘要:

Diffusion-based image super-resolution (SR) aims to reconstruct high-resolution (HR) images from low-resolution (LR) observations. A key property of diffusion models is that, once the starting point of the reverse chain and the denoising network are fixed, the quality of the generated image is dictated by the noise maps sampled at the intermediate steps, which are drawn from an unconstrained standard Gaussian distribution in conventional pipelines. This property raises a natural question: does there exist a noise sampling distribution better than the standard Gaussian that improves the quality of the generated images? To this end, we use a parameterized deep neural network to predict the mean and the variance of the Gaussian noise sampling distribution at each intermediate step, and design two training schemes, one supervised by the quality of the final generated image and the other by aligning each reverse step with the forward posterior. Experiments show that, compared with the standard Gaussian distribution, the learnable noise sampling distribution improves the quality of the generated images. The source code of our method can be found at https://github.com/Faze-Hsw/LPNSR.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑