WPPNets and WPPFlows: The Power of Wasserstein Patch Priors for Superresolution
From MaRDI portal
(Redirected from Publication:6168325)
Optimal transportation (49Q22) Bayesian problems; characterization of Bayes procedures (62C10) Bayesian inference (62F15) Artificial neural networks and deep learning (68T07) Reasoning under uncertainty in the context of artificial intelligence (68T37) Computing methodologies for image processing (68U10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)
Abstract: Exploiting image patches instead of whole images have proved to be a powerful approach to tackle various problems in image processing. Recently, Wasserstein patch priors (WPP), which are based on the comparison of the patch distributions of the unknown image and a reference image, were successfully used as data-driven regularizers in the variational formulation of superresolution. However, for each input image, this approach requires the solution of a non-convex minimization problem which is computationally costly. In this paper, we propose to learn two kind of neural networks in an unsupervised way based on WPP loss functions. First, we show how convolutional neural networks (CNNs) can be incorporated. Once the network, called WPPNet, is learned, it can be very efficiently applied to any input image. Second, we incorporate conditional normalizing flows to provide a tool for uncertainty quantification. Numerical examples demonstrate the very good performance of WPPNets for superresolution in various image classes even if the forward operator is known only approximately.
Recommendations
- Wasserstein loss for image synthesis and restoration
- Multi-frame super resolution via deep plug-and-play CNN regularization
- Wasserstein generative models for patch-based texture synthesis
- Enhanced image prior for unsupervised remoting sensing super-resolution
- The Little Engine that Could: Regularization by Denoising (RED)
Cites work
- A generative model for texture synthesis based on optimal transport between feature distributions
- A nonlocal Bayesian image denoising algorithm
- A Review of Image Denoising Algorithms, with a New One
- Accelerating GMM-Based Patch Priors for Image Restoration: Three Ingredients for a <inline-formula> <tex-math notation="LaTeX">$100{\times}$ </tex-math> </inline-formula> Speed-Up
- Compressive learning for patch-based image denoising
- Computed tomography reconstruction using deep image prior and learned reconstruction methods
- Convolutional proximal neural networks and plug-and-play algorithms
- Cubic convolution interpolation for digital image processing
- FSIM: A Feature Similarity Index for Image Quality Assessment
- General state space Markov chains and MCMC algorithms
- Generalized Normalizing Flows via Markov Chains
- High-dimensional mixture models for unsupervised image denoising (HDMI)
- Invertible neural networks versus MCMC for posterior reconstruction in grazing incidence X-ray fluorescence
- On the Douglas-Rachford splitting method and the proximal point algorithm for maximal monotone operators
- Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
- PCA reduced Gaussian mixture models with applications in superresolution
- Proximal Splitting Algorithms for Convex Optimization: A Tour of Recent Advances, with New Twists
- Shared Prior Learning of Energy-Based Models for Image Reconstruction
- Signal Recovery by Proximal Forward-Backward Splitting
- Single Image Super Resolution using a Joint GMM Method
- Stochastic normalizing flows for inverse problems: a Markov chains viewpoint
- The Little Engine that Could: Regularization by Denoising (RED)
- Variational networks: an optimal control approach to early stopping variational methods for image restoration
- Wasserstein generative models for patch-based texture synthesis
Cited in
(6)- Wasserstein loss for image synthesis and restoration
- Image super-resolution with PCA reduced generalized Gaussian mixture models in materials science
- Learning from small data sets: patch-based regularizers in inverse problems for image reconstruction
- Robustness and exploration of variational and machine learning approaches to inverse problems: an overview
- Semi-unbalanced optimal transport for reference-based image restoration and synthesis
- Stochastic super-resolution for Gaussian microtextures
This page was built for publication: WPPNets and WPPFlows: The Power of Wasserstein Patch Priors for Superresolution
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6168325)
