Gabor is Enough: Interpretable Deep Denoising with a Gabor Synthesis Dictionary Prior

04/23/2022
by   Nikola Janjušević, et al.
0

Image processing neural networks, natural and artificial, have a long history with orientation-selectivity, often described mathematically as Gabor filters. Gabor-like filters have been observed in the early layers of CNN classifiers and even throughout low-level image processing networks. In this work, we take this observation to the extreme and explicitly constrain the filters of a natural-image denoising CNN to be learned 2D real Gabor filters. Surprisingly, we find that the proposed network (GDLNet) can achieve near state-of-the-art denoising performance amongst popular fully convolutional neural networks, with only a fraction of the learned parameters. We further verify that this parameterization maintains the noise-level generalization (training vs. inference mismatch) characteristics of the base network, and investigate the contribution of individual Gabor filter parameters to the performance of the denoiser. We present positive findings for the interpretation of dictionary learning networks as performing accelerated sparse-coding via the importance of untied learned scale parameters between network layers. Our network's success suggests that representations used by low-level image processing CNNs can be as simple and interpretable as Gabor filterbanks.

READ FULL TEXT

page 1

page 4

research
12/02/2021

CDLNet: Noise-Adaptive Convolutional Dictionary Learning Network for Blind Denoising and Demosaicing

Deep learning based methods hold state-of-the-art results in low-level i...
research
03/06/2018

Nonlocality-Reinforced Convolutional Neural Networks for Image Denoising

We introduce a paradigm for nonlocal sparsity reinforced deep convolutio...
research
03/05/2021

CDLNet: Robust and Interpretable Denoising Through Deep Convolutional Dictionary Learning

Deep learning based methods hold state-of-the-art results in image denoi...
research
12/14/2016

Sparse Factorization Layers for Neural Networks with Limited Supervision

Whereas CNNs have demonstrated immense progress in many vision problems,...
research
11/22/2018

Three-dimensional Optical Coherence Tomography Image Denoising via Multi-input Fully-Convolutional Networks

In recent years, there has been a growing interest in applying convoluti...
research
09/12/2019

Rethinking the CSC Model for Natural Images

Sparse representation with respect to an overcomplete dictionary is ofte...
research
12/30/2019

Recognizing Instagram Filtered Images with Feature De-stylization

Deep neural networks have been shown to suffer from poor generalization ...

Please sign up or login with your details

Forgot password? Click here to reset