SVNR: Spatially-variant Noise Removal with Denoising Diffusion

TL;DR

SVNR introduces a spatially-variant noise model for diffusion-based image denoising, effectively handling realistic noise patterns by conditioning the process on the input image and allowing pixel-wise time embeddings.

Contribution

It proposes a novel diffusion formulation that models spatially-variant noise and incorporates the input image as a condition, improving denoising performance on real-world noise.

Findings

Outperforms baseline diffusion models in denoising tasks.

Achieves superior results compared to state-of-the-art single image denoising methods.

Demonstrates robustness to realistic, spatially-varying noise patterns.

Abstract

Denoising diffusion models have recently shown impressive results in generative tasks. By learning powerful priors from huge collections of training images, such models are able to gradually modify complete noise to a clean natural image via a sequence of small denoising steps, seemingly making them well-suited for single image denoising. However, effectively applying denoising diffusion models to removal of realistic noise is more challenging than it may seem, since their formulation is based on additive white Gaussian noise, unlike noise in real-world images. In this work, we present SVNR, a novel formulation of denoising diffusion that assumes a more realistic, spatially-variant noise model. SVNR enables using the noisy input image as the starting point for the denoising diffusion process, in addition to conditioning the process on it. To this end, we adapt the diffusion process to allow each pixel to have its own time embedding, and propose training and inference schemes that support spatially-varying time maps. Our formulation also accounts for the correlation that exists between the condition image and the samples along the modified diffusion process. In our experiments we demonstrate the advantages of our approach over a strong diffusion model baseline, as well as over a state-of-the-art single image denoising method.