Image Denoising Via Learned Dictionaries and Sparse representation

By Michael Elad, Michal Aharon. Computer Vision and Pattern Recognition, 2006

451 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/1827aff2dc345680de29a937f002da41accb2fe0

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 Media Technology

Citations in the corpus

Fast Inference in Sparse Coding Algorithms with Applications to Object Recognition (2010, background): “It has been shown that making these representations adaptive to image data can improve perform ance on vision tasks as demonstrated in [1] in a supervised learning framework and in [2, 3] using u ns pervised learning.”
Learning Multiscale Sparse Representations for Image and Video Restoration (2008, methodology): “Our results for denoising grayscale images outperform for instance works such as [10, 19, 20, 22, 35, 38].”
From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images (2009, methodology): “In fact, papers [63, 64] apply the K-SVD algorithm as shown in Exhibit 4 to image patches carved out of the noisy image.”
Multiresolution Bilateral Filtering for Image Denoising (2008, unknown): “[ 10 ]‐[12] develop gray-scale and color image denoising algorithms based on sparse and redundant representations over learned dictionaries, where training and denoising use the K-SVD algorithm; [13] and [14] group 2-D image fragments into 3-D data arrays, and apply a collaborative filtering procedu”
Sparse Representation for Signal Classification (2006, methodology): “i=1 ‖yi − Axi‖22 robustly recover the signal structure, as in [9, 18].”
On Single Image Scale-Up Using Sparse-Representations (2010, methodology): “The global operator Lall in this experiment is implemented by first blurring the high-resolution images y h with a 1D filter [1, 3, 4, 3, 1]/12 both horizontally and vertically, and then down-sampling it by a factor s = 3.”
A Unified Energy-Based Framework for Unsupervised Learning (2007, background): “Energybased approaches produce suitable energy surfaces by minimizing other loss functionals than the negative log likelihood (see [1] for a general introduction to energy-based learning in the supervised case).”
Towards Robust Blind Face Restoration with Codebook Lookup Transformer (2022, unknown): “Sparse representation with learned dictionaries has demonstrated its superiority in image restoration tasks, such as super-resolution [12, 31, 32, 39] and denoising [9].”
On the use of deep learning for computational imaging (2020, background): “) Alternatively, S may be learnt from data that are statistically similar to typical objects, if available; it is then referred to as a dictionary.(22,23)”
Learning to Sense Sparse Signals: Simultaneous Sensing Matrix and Sparsifying Dictionary Optimization (2009, methodology): “In fact, both the overcomplete dictionary, , and projection matrix, , are learned from the dataset (see next section for the general case of simultaneous learning).”
Image Sequence Denoising via Sparse and Redundant Representations (2009, methodology): “The basic assumption of the denoising method developed in [1] and [2] is that each image patch (of a fixed size) can be represented as a linear combination of a small subset of patches (atoms), taken from a fixed dictionary.”
Fields of Experts (2009, background): “Modeling entire images has also been considered in the context of image denoising (Elad and Aharon 2006).”
Sparse Representation for Color Image Restoration (2008, methodology): “The gray-scale K-SVD algorithm introduced in [3], [ 2 ]”
Large Scale Spectral Clustering Via Landmark-Based Sparse Representation (2015, methodology): “In the last decades, sparse coding [9], [32] method is proposed as a matrix factorization technique to model the human visual cortex, and later has been shown useful for applications in many fields such as visual neuroscience [22] and image restoration [12], [30].”
Restoring an Image Taken through a Window Covered with Dirt or Rain (2013, background): “Removing localized corruption can be considered a form of blind inpainting, where the position of the corrupted regions is not given (unlike traditional inpainting [5]).”