Explainable bilevel optimization: an application to the Helsinki Deblur Challenge

DOI10.3934/IPI.2022055zbMATH Open1515.94013arXiv2210.10050MaRDI QIDQ6169501FDOQ6169501

Authors: S. Bonettini, Giorgia Franchini, M. Prato 

Publication date: 12 July 2023

Published in: Inverse Problems and Imaging (Search for Journal in Brave)

Abstract: In this paper we present a bilevel optimization scheme for the solution of a general image deblurring problem, in which a parametric variational-like approach is encapsulated within a machine learning scheme to provide a high quality reconstructed image with automatically learned parameters. The ingredients of the variational lower level and the machine learning upper one are specifically chosen for the Helsinki Deblur Challenge 2021, in which sequences of letters are asked to be recovered from out-of-focus photographs with increasing levels of blur. Our proposed procedure for the reconstructed image consists in a fixed number of FISTA iterations applied to the minimization of an edge preserving and binarization enforcing regularized least-squares functional. The parameters defining the variational model and the optimization steps, which, unlike most deep learning approaches, all have a precise and interpretable meaning, are learned via either a similarity index or a support vector machine strategy. Numerical experiments on the test images provided by the challenge authors show significant gains with respect to a standard variational approach and performances comparable with those of some of the proposed deep learning based algorithms which require the optimization of millions of parameters.

Full work available at URL: https://arxiv.org/abs/2210.10050

Recommendations

• A bilevel optimization approach for parameter learning in variational models
• Bilevel approaches for learning of variational imaging models
• Bilevel optimization for calibrating point spread functions in blind deconvolution
• Bilevel parameter learning for higher-order total variation regularisation models
• Inexact derivative-free optimization for bilevel learning

zbMATH Keywords

bilevel optimizationperformance predictorblind deconvolution methodsGreenAIHelsinki Deblur Challenge

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Learning and adaptive systems in artificial intelligence (68T05) Nonlinear programming (90C30) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Cites Work

• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• Smooth minimization of non-smooth functions
• The Little Engine that Could: Regularization by Denoising (RED)
• A bilevel optimization approach for parameter learning in variational models
• A scaled gradient projection method for constrained image deblurring
• On the convergence of the iterates of the ``fast iterative shrinkage/thresholding algorithm
• Insights Into Analysis Operator Learning: From Patch-Based Sparse Models to Higher Order MRFs
• Fields of experts
• Projected subgradient methods with non-Euclidean distances for non-differentiable convex minimization and variational inequalities
• Efficient gradient projection methods for edge-preserving removal of Poisson noise
• Inertial variable metric techniques for the inexact forward-backward algorithm
• Deep unfolding of a proximal interior point method for image restoration
• Solving inverse problems using data-driven models
• Bilevel optimization with nonsmooth lower level problems
• Computed tomography with view angle estimation using uncertainty quantification
• New convergence results for the scaled gradient projection method
• Variational networks: an optimal control approach to early stopping variational methods for image restoration
• Adaptive FISTA for Nonconvex Optimization
• Ritz-like values in steplength selections for stochastic gradient methods
• Neural architecture search via standard machine learning methodologies

Cited In (2)

• Bilevel optimal parameter learning for a high-order nonlocal multiframe super-resolution problem
• A new proximal heavy ball inexact line-search algorithm

This page was built for publication: Explainable bilevel optimization: an application to the Helsinki Deblur Challenge

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6169501)