Optimal Spatial Adaptation for Patch-Based Image Denoising

By Charles Kervrann, Jérôme Boulanger. IEEE Transactions on Image Processing, 2006

487 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/538282dfc40850df2fbfa7a6bbe8c274751b32f1

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Restoring Vision in Adverse Weather Conditions With Patch-Based Denoising Diffusion Models (2022, unknown): “An important drawback of this approach so far has been that the resulting image can contain merging artifacts from independently restored intermediate results, which was extensively studied in traditional restoration methods [63], [64], [65].”
Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering (2007, methodology): “As it is shown in the previous section, even with the currently used standard transforms, our algorithm already demonstrates better performance than both [8] and [11].”
Is Denoising Dead? (2010, background): “[ 2 ] improved on the performance of NLM by incorporating an adaptive search window along with a”
Patch Complexity, Finite Pixel Correlations and Optimal Denoising (2012, background): “This motivates an adaptive denoising scheme [12] where each pixel is denoised with a variable patch size that depends on its local image complexity.”
Patch-Based Nonlocal Functional for Denoising Fluorescence Microscopy Image Sequences (2010, background): “Since the bias is unknown, we consider instead a weaker “oracle” to detect the optimal bandwidth for smoothing (see [56] and [57])”
The SURE-LET Approach to Image Denoising (2007, background): “Several works along this line have been quite recently undertaken [24,25].”
Image Noise Level Estimation by Principal Component Analysis (2013, methodology): “We utilized the denoising method [2], which outperforms the methods [46], [47], [48], [49] and can be considered as the state of the art in image denoising.”
Poisson Noise Reduction with Non-local PCA (2012, methodology): “in our experiments). 7.4.2 Other methods We compare our method with other recent algorithms designed for retrieval of Poisson corrupted images. In the case of 2D images we have compared with: { SAFIR [20,5], using Anscombe transform and the rened inverse transform proposed in [28]. { Poisson multisc”
The Little Engine That Could: Regularization by Denoising (RED) (2016, unknown): “…algorithms published over the years, we obviously focus on the best performing ones, such as the nonlocal means (NLM) and its advanced variants [1, 2, 3], the K-SVD denoising method that relies on sparse representation modeling of image patches [4] and its nonlocal extension [7], the…”
An Optimized Blockwise Nonlocal Means Denoising Filter for 3-D Magnetic Resonance Images (2008, methodology): “More recently, some promising methods have been proposed for improved image denoising, based on statistical averaging schemes enhanced via incorporating a variable spatial neighborhood scheme [25](cid:150)[29].”
Sparse Modeling for Image and Vision Processing (2014, unknown): “First attempts have focused on improving the computational speed [Mahmoudi and Sapiro, 2005], or on automatically adapting the kernel bandwidth h to the pixel of interest [Kervrann and Boulanger, 2006].”
Clustering-Based Denoising With Locally Learned Dictionaries (2009, methodology): “where is the number of pixels in each patch, is a constant and is the standard deviation of the corrupting noise which is assumed to be known or can be estimated [ 4 ], [36].”
Two-stage image denoising by principal component analysis with local pixel grouping (2010, methodology): “This scheme was improved in [19], where the pair-wise hypothesis testing was used in the NLM estimation.”
Pointwise Shape-Adaptive DCT for High-Quality Denoising and Deblocking of Grayscale and Color Images (2007, unknown): “40 Patch-based (Kervrann and Boulanger) [29] 33.”
BM3D Image Denoising with Shape-Adaptive Principal Component Analysis (2009, background): “…regression [16] (also combined with learned dictionaries [2]); shape-adaptive DCT (SA-DCT) on neighborhoods whose shapes are adaptive to the image structures [7]; nonlocal (cid:222) ltering based on the assumption that there exist mutually similar blocks in natural images [1], [12], [3], [4].”
Learning Multiscale Sparse Representations for Image and Video Restoration (2008, background): “Our results for denoising grayscale images outperform, for instance, works such as [10, 19, 20, 22, 35, 38].”
A Tour of Modern Image Filtering: New Insights and Methods, Both Practical and Theoretical (2013, methodology): “The NLM algorithm [7], [26], [27], originally proposed in [28] and [29], has stirred a great deal of interest in the community in recent years.”