Non-local Regularization of Inverse Problems

By G. Peyr'e, Sébastien Bougleux, Laurent D. Cohen. European Conference on Computer Vision, 2008

383 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Mathematical Physics

View Paper: https://www.semanticscholar.org/paper/2f4d8b30592bffb8a5f122ba51547515f0627172

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Mathematical Physics

Citations in the corpus

A supervised patch-based approach for human brain labeling (2011, positive): “this nonlocal strategy has been studied and applied in several image processing applications such as nonlocal regularization functionals in the context of inverse problems [20], [27], [14], [29], [31] or medical image synthesis [32].”
Bregmanized Nonlocal Regularization for Deconvolution and Sparse Reconstruction (2010, methodology): “In [24], the authors have also proposed to update the graph weight to solve inverse problems using the forward-backward operator splitting technique [10].”
Accelerated filtering on graphs using Lanczos method (2015, background): “For instance, many applications in image processing beneﬁt from the use of graphs that describe connections between local patches in the image (e.g. [ 13, 22] and references therein).”
Single image super-resolution using Gaussian process regression (2011, methodology): “Previously, this self-similarity property has been exploited in image denoising by NonLocal Means [3], which was later generalized to a regularization method for inverse problems in [16].”
ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Measurements (2016, methodology): “In the second, the algorithms enforce the image model implicitly through a non-local regularization term in the objective function [26, 34, 7].”
From Denoising to Compressed Sensing (2014, background): “Through extensive simulations we show that in high-dimensional settings (for the subset of denoisers that we consider in this paper) our state evolution predicts the mean square error (MSE) of D-AMP accurately.”
Patch Group Based Nonlocal Self-Similarity Prior Learning for Image Denoising (2015, methodology): “The nonlocal means [11] and nonlocal regularization [29] methods improve much the image denoising performance over the conventional local selfsimilarity based methods.”
Recovering missing pixels for Landsat ETM + SLC-off imagery using multi-temporal regression analysis and a regularization method (2013, positive): “Several studies (Lou et al., 2010; Peyré et al., 2008) have shown that using the constraint of a non-local regularization method can effectively improve the detailed texture of the processing results.”
Super-resolution: a comprehensive survey (2014, background): “– Non-local graph-based regularization [358].”
An introduction to continuous optimization for imaging (2016, background): “These approaches are not variational at first glance, but variants can be obtained by alternating minimization of non-local energies (Peyré et al. 2008, Arias et al. 2011).”
Bilevel approaches for learning of variational imaging models (2015, methodology): “In the context of data driven learning approaches we mention statistically grounded methods for optimal model design [50] and marginalization [14, 62], adaptive and multiscale regularization [94, 41, 45] – also for non-local regularisation [13, 47, 80, 81] – learning in the context of sparse coding ”
Image denoising with nonlocal spectral graph wavelets (2010, methodology): “As a direction for future research, we may consider jointly estimating the graph weights and the denoised image, through a block relaxation scheme following [39].”
Image Recovery via Nonlocal Operators (2010, background): “The application of the nonlocal means filter to different image processing tasks is an active research area [4, 12, 22].”
Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights (2009, methodology): “In [44], the parameters are estimated iteratively by the maximum a posteriori (MAP) estimator under assumptions of interactions between pixels of the “true” image parameters.”