NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
This work provides a more computationally efficient and effective image denoising solution for applications requiring high-quality image reconstruction from noisy inputs, such as photography and medical imaging.
This paper introduces NBNet, a new image denoising framework that tackles noise reduction through image-adaptive projection. By learning reconstruction bases in the feature space and projecting input into the signal subspace, NBNet achieves state-of-the-art performance on SIDD and DND benchmarks in terms of PSNR and SSIM, with significantly reduced computational cost.
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.