K
Karen Egiazarian
Researcher at Tampere University of Technology
Publications - 603
Citations - 26910
Karen Egiazarian is an academic researcher from Tampere University of Technology. The author has contributed to research in topics: Image processing & Filter (signal processing). The author has an hindex of 53, co-authored 585 publications receiving 22477 citations. Previous affiliations of Karen Egiazarian include Nokia & Roma Tre University.
Papers
More filters
Journal ArticleDOI
Sparse approximations in complex domain based on BM3D modeling
TL;DR: Complex domain denoising is developed and studied as a test-problem for comparison of the designed filters as well as the different types of sparsity modeling.
Proceedings Article
Image upsampling via spatially adaptive block-matching filtering
TL;DR: A novel wavelet-domain image upsampling algorithm based on iterative spatially adaptive filtering and the Block-Matching and 3D filtering technique, which results in high-quality upsampled images, with sharp edges and practically no artifacts.
Journal ArticleDOI
Blind Source Separation by Entropy Rate Minimization
TL;DR: An algorithm for the blind separation of mutually independent and/or temporally correlated sources is presented in this letter and uses a simpler contrast function that can be accurately and efficiently estimated using nearest-neighbor distances.
Journal ArticleDOI
Efficiency of texture image enhancement by DCT-based filtering
Aleksey Rubel,Vladimir V. Lukin,Mikhail L. Uss,Benoit Vozel,Oleksiy Pogrebnyak,Karen Egiazarian +5 more
TL;DR: It is shown that noise removal in texture images using the considered techniques can distort fine texture details, and filtering efficiency predictors, including neural network based predictor, applicable to a wide class of images are proposed.
Pointwise shape-adaptive dct as an overcomplete denoising tool
TL;DR: A novel approach to image-denoising based on the shapeadaptive DCT (SA-DCT) is presented: the anisotropic LPAICI technique is used in order to de-shaped the shape of the transforms support in a pointwise adaptive manner.