下一尺度预测：一种用于真实世界图像去噪的自监督方法
Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising
提出跨尺度训练对构建方式，将噪声去相关与细节保留解耦，可推广到其他自监督底层视觉任务（如去模糊、超分）中，避免像素洗牌下采样对结构的破坏。
原文依据 NSP constructs cross-scale training pairs, where BSN takes low-resolution, fully decorrelated sub-images as input to predict high-resolution targets that retain fine details.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论自监督真实世界图像去噪仍是一项基本挑战，源于对空间结构噪声去相关与保留高频细节之间的对抗性权衡。现有的盲点网络（BSN）方法依赖像素洗牌下采样（PD）来对噪声去相关，但激进的下采样会碎片化精细结构，而较温和的下采样则无法去除相关噪声。为解决这一问题，我们引入下一尺度预测（NSP），一种新颖的自监督范式，将噪声去相关与细节保留解耦。NSP构建跨尺度训练对，其中BSN以低分辨率、完全去相关的子图像作为输入，预测保留精细细节的高分辨率目标。作为副产品，NSP无需重新训练或修改即可自然支持噪声图像的超分辨率。大量实验表明，NSP在真实世界基准上达到了最优的自监督去噪性能，显著缓解了噪声去相关与细节保留之间长期存在的冲突。代码可在 https://github.com/XLearning-SCU/2026-CVPR-NSP 获取。
英文原文摘要 1171 字符
Self-supervised real-world image denoising remains a fundamental challenge, arising from the antagonistic trade-off between decorrelating spatially structured noise and preserving high-frequency details. Existing blind-spot network (BSN) methods rely on pixel-shuffle downsampling (PD) to decorrelate noise, but aggressive downsampling fragments fine structures, while milder downsampling fails to remove correlated noise. To address this, we introduce Next-Scale Prediction (NSP), a novel self-supervised paradigm that decouples noise decorrelation from detail preservation. NSP constructs cross-scale training pairs, where BSN takes low-resolution, fully decorrelated sub-images as input to predict high-resolution targets that retain fine details. As a by-product, NSP naturally supports super-resolution of noisy images without retraining or modification. Extensive experiments demonstrate that NSP achieves state-of-the-art self-supervised denoising performance on real-world benchmarks, significantly alleviating the long-standing conflict between noise decorrelation and detail preservation. The code is available at https://github.com/XLearning-SCU/2026-CVPR-NSP.