Fast and Interpretable Nonlocal Neural Networks for Image Denoising via Group-Sparse Convolutional Dictionary Learning

2 Jun 2023  ·  Nikola Janjušević, Amirhossein Khalilian-Gourtani, Adeen Flinker, Yao Wang ·

Nonlocal self-similarity within natural images has become an increasingly popular prior in deep-learning models. Despite their successful image restoration performance, such models remain largely uninterpretable due to their black-box construction. Our previous studies have shown that interpretable construction of a fully convolutional denoiser (CDLNet), with performance on par with state-of-the-art black-box counterparts, is achievable by unrolling a dictionary learning algorithm. In this manuscript, we seek an interpretable construction of a convolutional network with a nonlocal self-similarity prior that performs on par with black-box nonlocal models. We show that such an architecture can be effectively achieved by upgrading the $\ell 1$ sparsity prior of CDLNet to a weighted group-sparsity prior. From this formulation, we propose a novel sliding-window nonlocal operation, enabled by sparse array arithmetic. In addition to competitive performance with black-box nonlocal DNNs, we demonstrate the proposed sliding-window sparse attention enables inference speeds greater than an order of magnitude faster than its competitors.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Grayscale Image Denoising BSD68 sigma15 GroupCDL PSNR 31.82 # 8
Grayscale Image Denoising BSD68 sigma25 GroupCDL PSNR 29.38 # 5
Grayscale Image Denoising BSD68 sigma50 GroupCDL PSNR 26.47 # 5


No methods listed for this paper. Add relevant methods here