Video Deblurring Algorithm Using Accurate Blur Kernel Estimation and Residual Deconvolution Based on a Blurred-Unblurred Frame Pair

2013 ◽  
Vol 22 (3) ◽  
pp. 926-940 ◽  
Author(s):  
Dong-Bok Lee ◽  
Shin-Cheol Jeong ◽  
Yun-Gu Lee ◽  
Byung Cheol Song
Keyword(s):  
Kernel Estimation ◽  
Blur Kernel Estimation ◽  
Blur Kernel

scholarly journals Blur kernel estimation of noisy-blurred image via dynamic structure prior

2020 ◽  
Vol 403 ◽  
pp. 268-281
Author(s):  
Xueling Chen ◽  
Yu Zhu ◽  
Wei Liu ◽  
Jinqiu Sun ◽  
Yanning Zhang
Keyword(s):  
Kernel Estimation ◽  
Dynamic Structure ◽  
Blur Kernel Estimation ◽  
Blurred Image ◽  
Blur Kernel

scholarly journals Motion Blur Kernel Estimation via Deep Learning

2018 ◽  
Vol 27 (1) ◽  
pp. 194-205 ◽  
Author(s):  
Xiangyu Xu ◽  
Jinshan Pan ◽  
Yu-Jin Zhang ◽  
Ming-Hsuan Yang
Keyword(s):  
Deep Learning ◽  
Kernel Estimation ◽  
Motion Blur ◽  
Blur Kernel Estimation ◽  
Blur Kernel

Enhancing Blurred Low-Resolution Images via Exploring the Potentials of Learning-Based Super-Resolution

2019 ◽  
Vol 33 (07) ◽  
pp. 1940007 ◽  
Author(s):  
Wen-Ze Shao ◽  
Bing-Kun Bao ◽  
Hai-Bo Li
Keyword(s):  
Kernel Estimation ◽  
Super Resolution ◽  
Energy Functional ◽  
Local Linear Regression ◽  
Estimation Accuracy ◽  
Deep Convolutional Neural Networks ◽  
Candidate Solution ◽  
Blur Kernel Estimation ◽  
Blur Kernel ◽  
Sr Method

This paper aims to propose a candidate solution to the challenging task of single-image blind super-resolution (SR), via extensively exploring the potentials of learning-based SR schemes in the literature. The task is formulated into an energy functional to be minimized with respect to both an intermediate super-resolved image and a nonparametric blur-kernel. The functional includes a so-called convolutional consistency term which incorporates a nonblind learning-based SR result to better guide the kernel estimation process, and a bi-[Formula: see text]-[Formula: see text]-norm regularization imposed on both the super-resolved sharp image and the nonparametric blur-kernel. A numerical algorithm is deduced via coupling the splitting augmented Lagrangian (SAL) and the conjugate gradient (CG) method. With the estimated blur-kernel, the final SR image is reconstructed using a simple TV-based nonblind SR method. The proposed blind SR approach is demonstrated to achieve better performance than [T. Michaeli and M. Irani, Nonparametric Blind Super-resolution, in Proc. IEEE Conf. Comput. Vision (IEEE Press, Washington, 2013), pp. 945–952.] in terms of both blur-kernel estimation accuracy and image ehancement quality. In the meanwhile, the experimental results demonstrate surprisingly that the local linear regression-based SR method, anchored neighbor regression (ANR) serves the proposed functional more appropriately than those harnessing the deep convolutional neural networks.

Sign in / Sign up

Export Citation Format

Share Document