2 Are built upon powerful patch-based (local) image models:  Non-Local Means (NLM): self-similarity within natural images  K-SVD: sparse representation modeling of image patches  BM3D: combines a sparsity prior and non local self-similarity  Kernel-regression: offers a local directional filter  EPLL: exploits a GMM model of the image patches  … Today we present a way to improve various su...

The paper is addressed to 2D phase and amplitude estimation of complex-valued signals —that is, in particular, to estimation of modulo-2π interferometric phase images from periodic and noisy observations. These degradation mechanisms make phase image estimation a challenging problem. A sparse nonlocal data-adaptive imaging formalized in complex domain is used for phase and amplitude image recon...

Generative adversarial networks (GANs) transform lowdimensional latent vectors into visually plausible images. If the real dataset contains only clean images, then ostensibly, the manifold learned by the GAN should contain only clean images. In this paper, we propose to denoise corrupted images by finding the nearest point on the GAN manifold, recovering latent vectors by minimizing distances i...

We consider the problem of the joint denoising of a number of rawdata images from a digital imaging sensor. In particular, we exploit a recently proposed image modeling [8] that incorporates both the signal-dependent nature of noise and the clipping of the data due to underor over-exposure of the sensor. Our denoising approach is based on the V-BM3D algorithm [5], coupled with a set of homomorp...

We propose a novel fast iterative thresholding algorithm for image compressive sampling (CS) recovery using three existing denoisers—i.e., TV (total variation), wavelet, and BM3D (block-matching and 3D filtering) denoisers. Through the use of the recently introduced plug-and-play prior approach, we turn these denoisers into CS solvers. Thus, our method can jointly utilize the global and nonloca...

Recent years have seen a surge of interest in deep neural networks fueled by their successful applications in numerous image processing and computer vision tasks. However, such applications typically come with huge computational loads. In this article, we explore the possibility of using small neural networks to denoise images. In particular, we present SSaNN (Self-Similarity and Neural Network...

In order to simultaneously sharpen image details and attenuate noise, we propose to combine the recent blockmatching and 3D Þltering (BM3D) denoising approach, based on 3D transform-domain collaborative Þltering, with alpha-rooting, a transform-domain sharpening technique. The BM3D exploits grouping of similar image blocks into 3D arrays (groups) on which collaborative Þltering (by hard-thresho...