Nonlocal Image Restoration With Bilateral Variance Estimation: A Low-Rank Approach

By Weisheng Dong, Guangming Shi, Xin Li. 2013

662 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computational Mechanics, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Biomedical Engineering

In Computational Mechanics

In Computer Vision and Pattern Recognition

Citations in the corpus

Weighted Nuclear Norm Minimization with Application to Image Denoising (2014, methodology): “Compared with WNNM, methods LSSC, NCSR and SAIST over-smooth more the textures in the sands area of image Boats , and methods BM3D and EPLL generate more artifacts.”
Weighted Schatten $p$ -Norm Minimization for Image Denoising and Background Subtraction (2015, methodology): “To restore clean image iteratively, we adopt the iterative regularization scheme in [7], which adds filtered residual back to the denoised image as follows: y(k) = x̂(k−1) + α(y − x̂(k−1)), (14)”
Variational Denoising Network: Toward Blind Noise Modeling and Removal (2019, methodology): “Instead of assuming i.i.d. distribution for the noise as conventional [28, 13, 16, 41], which largely deviates the spatial variant and signal-depend characteristics of the real noise [45, 8], we models the noise as a non-i.i.d. and pixel-wise Gaussian distribution in Eq.”
Brief review of image denoising techniques (2019, methodology): “Dong et al. [74] proposed a low-rank method based on SVD to model the sparse representation of non-locally similar image patches.”
Learning Enriched Features for Real Image Restoration and Enhancement (2020, background): “Numerous patch-based algorithms that exploit redundancy (selfsimilarity) in images are later developed [28,38,70,43].”
Weighted Nuclear Norm Minimization and Its Applications to Low Level Vision (2016, methodology): “The NSSbased image denoising algorithms such as BM3D (Dabov et al, 2007), LSSC (Mairal et al, 2009), NCSR (Dong et al, 2011) and SAIST (Dong et al, 2013) have achieved state-ofthe-art denoising results.”
Compressive Sensing via Nonlocal Low-Rank Regularization (2014, background): “Digital Object Identifier 10.1109/TIP.2014.2329449 I. INTRODUCTION THE theory of compressive sensing (CS) [1], [2]has attracted considerable research interests from signal/image processing communities.”
Towards Discriminability and Diversity: Batch Nuclear-Norm Maximization Under Label Insufficient Situations (2020, background): “reduce the influence of the extra components, minimizing nuclear-norm of the matrix has been applied to image denoising [15], image restoration [9] and many other tasks.”
Kronecker-Basis-Representation Based Tensor Sparsity and Its Applications to Tensor Recovery (2018, methodology): “…there are many matrix recovery methods constructed by adding more prior knowledge related to data cases together with matrix low-rankness [16], [22], such as joint trace/TV based MSI recovery [23], [27] and joint low-rank/sparsity based matrix recovery [55], [56], etc. Matrix completion…”
Denoising Prior Driven Deep Neural Network for Image Restoration (2018, background): “Motivated by the fact that natural images often contain rich repetitive structures, nonolocal regularization techniques [2], [4], [5], [8] combining the NLSS with the sparse representation and low-rank approximation, have shown signiﬁcant improvements over their local counterparts.”
Learning Enriched Features for Fast Image Restoration and Enhancement (2022, background): “Numerous patch-based schemes that exploit redundancy (self-similarity) in images are later developed [29], [30], [31], [32].”
Denoising of 3D magnetic resonance images by using higher-order singular value decomposition (2015, methodology): “The second stage in this proposed method is the HOSVD denoising of the combined image from original and pre-denoised images; this stage can also be considered a two-stage recursive regularization by adding filtered noise back to the denoised image (Dong et al., 2013; Osher et al., 2005): Yw 1⁄4 b XH”
LR3M: Robust Low-Light Enhancement via Low-Rank Regularized Retinex Model (2020, methodology): “In [34], a spatially adaptive iterative singular-value thresholding method is developed to restore images.”