TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Image Denoising	DND	NBNet	PSNR (sRGB)	39.89	# 4
Image Denoising	DND	NBNet	SSIM (sRGB)	0.955	# 5
Image Denoising	SIDD	NBNet	SSIM (sRGB)	0.973	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/nbnet-noise-basis-learning-for-image/image-denoising-on-sidd)](https://paperswithcode.com/sota/image-denoising-on-sidd?p=nbnet-noise-basis-learning-for-image)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/nbnet-noise-basis-learning-for-image/image-denoising-on-dnd)](https://paperswithcode.com/sota/image-denoising-on-dnd?p=nbnet-noise-basis-learning-for-image)`

NBNet: Noise Basis Learning for Image Denoising with Subspace Projection

CVPR 2021 · Shen Cheng, Yuzhi Wang, Haibin Huang, Donghao Liu, Haoqiang Fan, Shuaicheng Liu ·

In this paper, we introduce NBNet, a novel framework for image denoising. Unlike previous works, we propose to tackle this challenging problem from a new perspective: noise reduction by image-adaptive projection. Specifically, we propose to train a network that can separate signal and noise by learning a set of reconstruction basis in the feature space. Subsequently, image denosing can be achieved by selecting corresponding basis of the signal subspace and projecting the input into such space. Our key insight is that projection can naturally maintain the local structure of input signal, especially for areas with low light or weak textures. Towards this end, we propose SSA, a non-local subspace attention module designed explicitly to learn the basis generation as well as the subspace projection. We further incorporate SSA with NBNet, a UNet structured network designed for end-to-end image denosing. We conduct evaluations on benchmarks, including SIDD and DND, and NBNet achieves state-of-the-art performance on PSNR and SSIM with significantly less computational cost.