Nonlocally Centralized Sparse Representation for Image Restoration

By Weisheng Dong, Lei Zhang, Guangming Shi, Xin Li. 2013

1,468 citations recorded by the source corpus.

Subfields: 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 Computational Mechanics

In Computer Vision and Pattern Recognition

Citations in the corpus

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (2021, methodology): “As mentioned in the introduction, bicubic downsampling and traditional degradations underly existing DNNsbased SISR methods [22, 34, 41, 55], or some simple variants do [10, 11, 38, 47, 53, 54].”
Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising (2019, background): “Traditional image denoising methods generally focus on modeling natural image priors and use the priors to restore the clean image, including sparse models [5, 26], Markov random field models [13], etc.”
Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising (2016, methodology): “In particular, the NSS models are popular in state-of-the-art methods such as BM3D [2], LSSC [4], NCSR [6] and WNNM [13].”
Deep Network Interpolation for Continuous Imagery Effect Transition (2018, background): “However, a bank of discriminative CNN denoisers are required and the hyper-parameter selection in optimization is not a trivial task [4].”
Learning a Single Convolutional Super-Resolution Network for Multiple Degradations (2017, methodology): “For example, the method proposed in [57] can handle the widelyused Gaussian degradation as in [11].”
ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting (2023, background): “Traditional image SR methods primarily focus on designing more rational image priors based on our subjective knowledge, such as non-local similarity [26], low-rankness [27], sparsity [28, 29], and so on.”
Toward Convolutional Blind Denoising of Real Photographs (2018, methodology): “Beneﬁted from the realistic noise model and incorporation with real noisy images, our CBDNet achieves the highest PSNR/SSIM results, and slightly better than MCWNNM [59] and TWSC [58].”
Seven Ways to Improve Example-Based Single Image Super Resolution (2015, neutral): “[9], among others [37, 12, 38, 16].”
Blind Super-Resolution With Iterative Kernel Correction (2019, background): “In literature, the most widely adopted blur kernel is isotropic Gaussian blur kernel [8, 36, 39].”
The Little Engine That Could: Regularization by Denoising (RED) (2016, neutral): “Figures 8 and 9 visually compare the proposed method to P 3 and also to the state-of-the-art NCSR.”
Sparse Modeling for Image and Vision Processing (2014, methodology): “The terminology “hard” is defined in contrast to an important variant called the “soft-thresholding operator”, which was introduced by Donoho and Johnstone [1994] in the context of wavelets:3 3Note that the soft-thresholding operator appears in fact earlier in the statistics literature [see Efron an”
Trainable Nonlinear Reaction Diffusion: A Flexible Framework for Fast and Effective Image Restoration (2015, methodology): “restoration are non-local methods based on patch modeling, for example, image denoising with (i) Gaussian noise [15], [26], [40], [46], (ii) multiplicative noise [14], or (iii) Poisson noise [24], image interpolation [47], image deconvolution [20], etc.”
Deep Unfolding Network for Image Super-Resolution (2020, methodology): “In [15, 42], the degradation model focuses on Gaussian blurring and a subsequent downsampling with scale factor 3.”
Image Blind Denoising with Generative Adversarial Network Based Noise Modeling (2018, methodology): “NCSR and BM3D are evaluated at a default setting σ = 25.”
Image denoising using deep CNN with batch renormalization (2020, background): “Non-locally centralized sparse representation (NCSR) centralized the sparse coding to suppress the sparse-coding noise [14].”
Variational Denoising Network: Toward Blind Noise Modeling and Removal (2019, methodology): “Comparison methods: Several state-of-the-art denoising methods are adopted for performance comparison, including CBM3D [11], WNNM [16], NCSR [14], MLP [10], DnCNN-B [44], Mem-Net [38], FFDNet [45], UDNet [24] and CBDNet [17].”
Real Image Denoising With Feature Attention (2019, neutral): “8306 NCSR [23] Non-blind 34.”
A Statistical Prediction Model Based on Sparse Representations for Single Image Super-Resolution (2014, methodology): “– ASDS-AR, ASDS-AR-NL [12] and NCSR [22].”
CKD: Cross-Task Knowledge Distillation for Text-to-Image Synthesis (2020, neutral): “age super-resolution, sketch-photo synthesis [60]–[63].”