Learning Multiscale Sparse Representations for Image and Video Restoration

By Julien Mairal, Guillermo Sapiro, Michael Elad. Multiscale Modeling & simulation, 2008

525 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/323f70a2f5af24acb8feaca456da79d05d42b980

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

Citations in the corpus

Image Denoising Using the Higher Order Singular Value Decomposition (2013, unknown): “A multiscale variant of this algorithm (known as MS-KSVD) learns dictionaries to represent patches at two or more scales leading to further redundancy [20].”
Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization (2010, methodology): “Various extensions and variants of the K-SVD algorithm [27], [29]–[31] have been proposed to learn a universal and overcomplete dictionary.”
Image Denoising Methods. A New Nonlocal Principle (2010, methodology): “This approach and variants applied to video have been considered in [58], [75], and [23].”
Spectral–Spatial Hyperspectral Image Classification via Multiscale Adaptive Sparse Representation (2014, methodology): “We should note that the multiscale scheme has been utilized in a sparse representation technique for restoration problems and a more accurately reconstructed image has been obtained [37], [38].”
Sparse Modeling for Image and Vision Processing (2014, methodology): “To deal with unobserved information, the dictionary learning formulation can be modified by introducing a binary mask Mi for every patch indexed by i [Mairal et al., 2008c,e].”
A Comprehensive Survey to Face Hallucination (2013, methodology): “In some applications, such as image denoising (Elad and Aharon 2006) and image restoration (Mairal et al. 2008a), the dictionary is always learned from training examples by alternatively optimizing D and xcoe respectively, while in applications such as face recognition (Wright et al. 2009), D is…”
A Survey of Sparse Representation: Algorithms and Applications (2015, methodology): “[175] presented a multiscale sparse image representation framework based on the KSVD dictionary learning algorithm and shift-invariant sparsity prior knowledge for restoration of color images and video image sequence.”
MR Image Reconstruction From Highly Undersampled k-Space Data by Dictionary Learning (2011, methodology): “Dictionaries adapted to the specific image that is to be recovered have been shown to provide state of the art results in image denoising [ 30 ].”
Image super-resolution as sparse representation of raw image patches (2008, background): “Recently sparse representation has been applied to many other related inverse problems in image processing, such as compression, denoising [10], and restoration [17], often improving on the state-of-the-art.”
Nonparametric Bayesian Dictionary Learning for Analysis of Noisy and Incomplete Images (2011, methodology): “Many of the existing methods for learning dictionaries are based on solving an optimization problem [1], [13], [24]–[26], [27], [28], in which one seeks to match the dictionary to”
Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation (2014, methodology): “In this paper, we develop a half-quadratic (HQ) framework to solve the robust sparse representation problem.”
From Local Kernel to Nonlocal Multiple-Model Image Denoising (2009, methodology): “[75] , the TLS algorithm by Hirakawa and Parks [47], the BLS-GSM algorithm by Portilla et al.”
Patch-Based Nonlocal Functional for Denoising Fluorescence Microscopy Image Sequences (2010, methodology): “Unlike the previous patch-based approaches [27], [31], [34], [36], we present in this paper a space-time patch-based adaptive statistical method for 3-D+time video-microscopy image sequence restoration.”
Multi-scale dictionary for single image super-resolution (2012, methodology): “We employ a multiscale K-SVD algorithm proposed in [12] to learn such a dictionary from {z i } N i=1 such that each sample can be approximately represented by a sparsely linear combination of the atoms from D, i.e., D, {α i } = arg min D,{αi} i z i − Dα i 2 2 , s.t.α i 0 ≤ L, ∀i (2) where α i is…”
Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity (2010, background): “, [44], an image is decompo sed into 128×128 regions, each region”
Sparse Representation for Computer Vision and Pattern Recognition (2010, methodology): “…variations and extensions of 1 minimization have been applied to many vision tasks, including face recognition [71], image super-resolution [75], motion and data segmentation [33], [56], supervised denoising and inpainting [51] and background modeling [16], [21] and image classification [47], [48].”
Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights (2009, background): “and sparse decompositions with over-complete or learned dictionaries [15]–[17].”