A convex variational model for learning convolutional image atoms from incomplete data

Abstract

A variational model for learning convolutional image atoms from corrupted and/or incomplete data is introduced and analyzed both in function space and numerically. Building on lifting and relaxation strategies, the proposed approach is convex and allows for simultaneous image reconstruction and atom-learning in a general, inverse problems context. Further, motivated by an improved numerical performance, also a semi-convex variant is included in the analysis and the experiments of the paper. For both settings, fundamental analytical properties allowing in particular to ensure well-posedness and stability results for inverse problems are proven in a continuous setting. Exploiting convexity, globally optimal solutions are further computed numerically for applications with incomplete, noisy and blurry data and numerical results are shown.

References in corpus (9)

• The Convex Geometry of Linear Inverse Problems
• Complete Dictionary Recovery over the Sphere I: Overview and the Geometric Picture
• Convolutional Dictionary Learning: A Comparative Review and New Algorithms
• Inertial Proximal Alternating Linearized Minimization (iPALM) for Nonconvex and Nonsmooth Problems
• Guaranteed Blind Sparse Spikes Deconvolution via Lifting and Convex Optimization
• Multi-Layer Convolutional Sparse Modeling: Pursuit and Dictionary Learning
• Inverse problems with second-order Total Generalized Variation constraints
• Tight convex relaxations for sparse matrix factorization
• A convex variational model for learning convolutional image atoms from incomplete data

Cited by in corpus (3)

• A convex variational model for learning convolutional image atoms from incomplete data
• Regularization Graphs -- A unified framework for variational regularization of inverse problems
• Deep learning based dictionary learning and tomographic image reconstruction