LoTA-N2N: Local Trace Adaptation for Zero-Shot Self-Supervised Image Denoising
2026-07-27 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors study how to improve denoising of images using only noisy images without having clean versions to compare to. They analyze why existing self-supervised methods sometimes struggle, especially when noise behaves unevenly or is unknown. They propose a new two-step method called LoTA-N2N that locally estimates and reduces errors caused by noise interactions, avoiding problems from averaging effects across the whole image. Their experiments on various types of images show that this approach consistently improves denoising quality under different noise conditions without needing extra clean data or noise models.
self-supervised denoisingmean squared error (MSE)noise correlationspatially nonstationary noiseblind-spot methodszero-shot adaptationimage denoisingnoise residuallocal interactionsspatial cancellation
Authors
Jintong Hu, Bin Xia, Junlin Liu, Jiayue Liu, Wenming Yang
Abstract
Single-image self-supervised denoising replaces unavailable clean targets with surrogate targets constructed from noisy observations. Its effectiveness therefore depends on how closely the surrogate objective remains aligned with supervised denoising, especially when noise is correlated, spatially nonstationary, or unknown. We express the discrepancy between a broad class of MSE-based self-supervised objectives and supervised MSE as a parameter-independent constant and a trace interaction between the surrogate-target residual and the prediction error. The corresponding gradient discrepancy is determined by the gradient of this interaction. This formulation provides a common view of paired-noise, blind-spot, weak-noise, re-corruption, and sub-image methods, while revealing that a small global interaction may conceal substantial positive and negative regional interactions through spatial cancellation. Building on these observations, we propose LoTA-N2N, a two-stage zero-shot adaptation framework. Stage 1 trains a denoiser on complementary sub-image pairs and freezes it to construct detached clean-sub-image proxies. Stage 2 estimates the residual--prediction interaction using these proxies and suppresses its patch-wise absolute magnitude. We show that the local construction prevents spatial cancellation and upper-bounds the magnitude of the corresponding global interaction. Experiments across natural, confocal, and X-ray images, complemented by iteration-matched controls, controlled noise shifts, and gradient diagnostics, show consistent gains over MSE-only adaptation under IID, spatially varying, and mixed noise. Overall, LoTA-N2N demonstrates that estimated local interaction and spatial cancellation control provide effective design principles for single-image self-supervised denoising without paired clean targets, repeated acquisitions, or a predefined re-corruption model.