Bilevel approaches for learning of variational imaging models
From MaRDI portal
Publication:4556138
Variational inequalities (49J40) Database theory (68P15) Learning and adaptive systems in artificial intelligence (68T05) Computing methodologies for image processing (68U10) Methods of quasi-Newton type (90C53) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)
Abstract: We review some recent learning approaches in variational imaging, based on bilevel optimisation, and emphasize the importance of their treatment in function space. The paper covers both analytical and numerical techniques. Analytically, we include results on the existence and structure of minimisers, as well as optimality conditions for their characterisation. Based on this information, Newton type methods are studied for the solution of the problems at hand, combining them with sampling techniques in case of large databases. The computational verification of the developed techniques is extensively documented, covering instances with different type of regularisers, several noise models, spatially dependent weights and large image databases.
Recommendations
- A bilevel optimization approach for parameter learning in variational models
- Inexact derivative-free optimization for bilevel learning
- Optimality Conditions for Bilevel Imaging Learning Problems with Total Variation Regularization
- Bilevel parameter learning for higher-order total variation regularisation models
- Bilevel optimization with nonsmooth lower level problems
Cited in
(57)- Variational networks: an optimal control approach to early stopping variational methods for image restoration
- A convex variational model for learning convolutional image atoms from incomplete data
- Automated parameter selection for total variation minimization in image restoration
- Inexact derivative-free optimization for bilevel learning
- Bilinear constraint based ADMM for mixed Poisson-Gaussian noise removal
- Adaptive image processing: a bilevel structure learning approach for mixed-order total variation regularizers
- A non-convex denoising model for impulse and Gaussian noise mixture removing using bi-level parameter identification
- An improved bilevel optimization approach for image super-resolution based on a fractional diffusion tensor
- An improved spatially controlled reaction-diffusion equation with a non-linear second order operator for image super-resolution
- A bilevel optimization approach for parameter learning in variational models
- Infimal convolution of data discrepancies for mixed noise removal
- Bilevel optimization with nonsmooth lower level problems
- A bilevel learning approach for optimal observation placement in variational data assimilation
- A high order PDE-constrained optimization for the image denoising problem
- Regularization graphs -- a unified framework for variational regularization of inverse problems
- Optimality Conditions for Bilevel Imaging Learning Problems with Total Variation Regularization
- Bilevel Training Schemes in Imaging for Total Variation--Type Functionals with Convex Integrands
- A Generative Variational Model for Inverse Problems in Imaging
- Dualization and Automatic Distributed Parameter Selection of Total Generalized Variation via Bilevel Optimization
- Convergence of a piggyback-style method for the differentiation of solutions of standard saddle-point problems
- Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks
- A nonsmooth trust-region method for locally Lipschitz functions with application to optimization problems constrained by variational inequalities
- Higher-order total variation approaches and generalisations
- Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. I: Methodology and experiments
- Learning regularization parameters of inverse problems via deep neural networks
- Gradient-based solution algorithms for a class of bilevel optimization and optimal control problems with a nonsmooth lower level
- Solving inverse problems using data-driven models
- Analysis and automatic parameter selection of a variational model for mixed Gaussian and salt-and-pepper noise removal
- Bilevel Methods for Image Reconstruction
- Bilevel inverse problems in neuromorphic imaging
- Bilevel Imaging Learning Problems as Mathematical Programs with Complementarity Constraints: Reformulation and Theory
- Higher order Ambrosio-Tortorelli scheme with non-negative spatially dependent parameters
- Bilevel optimal parameter learning for a high-order nonlocal multiframe super-resolution problem
- Learning Regularization Parameter-Maps for Variational Image Reconstruction Using Deep Neural Networks and Algorithm Unrolling
- Explainable bilevel optimization: an application to the Helsinki Deblur Challenge
- On and Beyond Total Variation Regularization in Imaging: The Role of Space Variance
- Parameter space study of optimal scale-dependent weights in TV image denoising
- A Flexible Space-Variant Anisotropic Regularization for Image Restoration with Automated Parameter Selection
- Learning Variational Models with Unrolling and Bilevel Optimization
- Learning sparsity-promoting regularizers using bilevel optimization
- Generative models and Bayesian inversion using Laplace approximation
- Bilevel optimization methods in imaging
- Interpretable model learning in variational imaging: a bilevel optimization approach
- Learning nonlocal weights for second-order nonlocal super-resolution
- Learning sparsity-promoting regularizers for linear inverse problems
- Bilevel parameter learning for higher-order total variation regularisation models
- Group sparse representation-based color image restoration under mixed random-valued impulse and Gaussian noise
- Automatic nonstationary anisotropic Tikhonov regularization through bilevel optimization
- Single-loop methods for bilevel parameter learning in inverse imaging
- An adaptively inexact method for bilevel learning using primal-dual-style differentiation
- Deep unrolling for learning optimal spatially varying regularisation parameters for total generalised variation
- Nested Bregman iterations for decomposition problems
- Efficient nonlocal linear image denoising: bilevel optimization with nonequispaced fast Fourier transform and matrix-free preconditioning
- Data-driven methods for quantitative imaging
- Dyadic partition-based training schemes for TV/TGV denoising
- Enhancing deblurring through optimal controlled bilevel with nonlocal weighted model
- Algorithmic unfolding for image reconstruction and localization problems in single-molecule fluorescence microscopy
This page was built for publication: Bilevel approaches for learning of variational imaging models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4556138)
