Published on Tue May 09 2017

Adaptive Regularization of Some Inverse Problems in Image Analysis

Byung-Woo Hong, Ja-Keoung Koo, Martin Burger, Stefano Soatto

adaptive regularization scheme for optimizing composite energy functionals arising in image analysis problems. Scheme automatically trades off data fidelity and regularization depending on the current data fit.

0
0
0
Abstract

We present an adaptive regularization scheme for optimizing composite energy functionals arising in image analysis problems. The scheme automatically trades off data fidelity and regularization depending on the current data fit during the iterative optimization, so that regularization is strongest initially, and wanes as data fidelity improves, with the weight of the regularizer being minimized at convergence. We also introduce the use of a Huber loss function in both data fidelity and regularization terms, and present an efficient convex optimization algorithm based on the alternating direction method of multipliers (ADMM) using the equivalent relation between the Huber function and the proximal operator of the one-norm. We illustrate and validate our adaptive Huber-Huber model on synthetic and real images in segmentation, motion estimation, and denoising problems.

Thu Sep 08 2016
Computer Vision
Adaptive Regularization in Convex Composite Optimization for Variational Imaging Problems
An algorithm to determine the relative weight between data fidelity and regularization based on the residual. We consider a number of imaging problems including denoising, segmentation and motion estimation. We demonstrate the robustness and effectiveness of our adaptive regularization scheme.
0
0
0
Wed Jun 17 2009
Computer Vision
Adaptive Regularization of Ill-Posed Problems: Application to Non-rigid Image Registration
We introduce an adaptive regularization approach. We constrain the prior distribution to be aGauss-Markov random field. We apply this approach to non-rigid image registration to estimate the spatial transformation between two images.
0
0
0
Mon Jul 07 2014
Machine Learning
Low Complexity Regularization of Linear Inverse Problems
Inverse problems and regularization theory is a central theme in contemporary communications. The goal is to reconstruct an unknown signal from partial indirect, and possibly noisy, measurements. This review is intended to serve as a one-stop shop toward the understanding of the theoretical properties of the problems.
0
0
0
Thu Dec 12 2019
Computer Vision
Variational Coupling Revisited: Simpler Models, Theoretical Connections, and Novel Applications
Coupling models are becoming increasingly popular in image analysis. They involve auxiliary variables, such that their energy split into multiple fractional steps that can be solved easier and more efficiently. We present a first order coupling model that performs classical TV regularisation.
0
0
0
Thu Jun 16 2016
Machine Learning
CLEAR: Covariant LEAst-square Re-fitting with applications to image restoration
In this paper, we propose a new framework to remove parts of the systematic errors affecting popular restoration algorithms. We emphasize the importance of preserving the Jacobian (w.r.t. the observed signal) of the original estimator. Then, we provide an approach that has a "
0
0
0
Fri Oct 14 2011
Computer Vision
Controlled Total Variation regularization for inverse problems
This paper provides a new algorithm for solving inverse problems, based on the minimization of the norm and on the control of the Total Variation.
0
0
0