Designing and Learning Trainable Priors with Non-Cooperative Games

06/26/2020
by   Bruno Lecouat, et al.
11

We introduce a general framework for designing and learning neural networks whose forward passes can be interpreted as solving convex optimization problems, and whose architectures are derived from an optimization algorithm. We focus on non-cooperative convex games, solved by local agents represented by the nodes of a graph and interacting through regularization functions. This approach is appealing for solving imaging problems, as it allows the use of classical image priors within deep models that are trainable end to end. The priors used in this presentation include variants of total variation, Laplacian regularization, sparse coding on learned dictionaries, and non-local self similarities. Our models are parameter efficient and fully interpretable, and our experiments demonstrate their effectiveness on a large diversity of tasks ranging from image denoising and compressed sensing for fMRI to dense stereo matching.

READ FULL TEXT

page 16

page 17

research
09/03/2022

A CNC approach for Directional Total Variation

The core of many approaches for the resolution of variational inverse pr...
research
10/21/2020

Unrolling of Deep Graph Total Variation for Image Denoising

While deep learning (DL) architectures like convolutional neural network...
research
07/19/2019

Deep Graph-Convolutional Image Denoising

Non-local self-similarity is well-known to be an effective prior for the...
research
10/09/2019

SNIDER: Single Noisy Image Denoising and Rectification for Improving License Plate Recognition

In this paper, we present an algorithm for real-world license plate reco...
research
11/18/2021

A Trainable Spectral-Spatial Sparse Coding Model for Hyperspectral Image Restoration

Hyperspectral imaging offers new perspectives for diverse applications, ...
research
02/24/2021

Hyperspectral Denoising Using Unsupervised Disentangled Spatio-Spectral Deep Priors

Image denoising is often empowered by accurate prior information. In rec...
research
08/21/2018

Faster PET Reconstruction with Non-Smooth Priors by Randomization and Preconditioning

Uncompressed clinical data from modern positron emission tomography (PET...

Please sign up or login with your details

Forgot password? Click here to reset