Towards Robust Image Denoising with Scale Equivariance
This addresses a generalization gap in image denoising for applications requiring robustness to diverse noise conditions, though it appears incremental as it builds on existing denoising concepts with novel components.
The paper tackles the problem of image denoising models struggling to generalize to out-of-distribution noise patterns, particularly spatially variant noise, by proposing a robust blind denoising framework with scale equivariance, which consistently outperforms state-of-the-art methods on synthetic and real-world benchmarks.
Despite notable advances in image denoising, existing models often struggle to generalize beyond in-distribution noise patterns, particularly when confronted with out-of-distribution (OOD) conditions characterized by spatially variant noise. This generalization gap remains a fundamental yet underexplored challenge. In this work, we investigate \emph{scale equivariance} as a core inductive bias for improving OOD robustness. We argue that incorporating scale-equivariant structures enables models to better adapt from training on spatially uniform noise to inference on spatially non-uniform degradations. Building on this insight, we propose a robust blind denoising framework equipped with two key components: a Heterogeneous Normalization Module (HNM) and an Interactive Gating Module (IGM). HNM stabilizes feature distributions and dynamically corrects features under varying noise intensities, while IGM facilitates effective information modulation via gated interactions between signal and feature paths. Extensive evaluations demonstrate that our model consistently outperforms state-of-the-art methods on both synthetic and real-world benchmarks, especially under spatially heterogeneous noise. Code will be made publicly available.