{"ID":2825434,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.21038","arxiv_id":"2512.21038","title":"Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising","abstract":"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.","short_abstract":"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 aggressi...","url_abs":"https://arxiv.org/abs/2512.21038","url_pdf":"https://arxiv.org/pdf/2512.21038v2","authors":"[\"Yiwen Shan\",\"Haiyu Zhao\",\"Peng Hu\",\"Xi Peng\",\"Yuanbiao Gou\"]","published":"2025-12-24T08:06:17Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":605663,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2825434,"paper_url":"https://arxiv.org/abs/2512.21038","paper_title":"Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising","repo_url":"https://github.com/XLearning-SCU/2026-CVPR-NSP","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}