Iterative Reweighted Image Inpainting Using Robustness to Outliers of L1 Norm Minimization

Tomohiro TAKAHASHI  Masaki NAKANISHI  Kazunori URUMA  Toshihiro FURUKAWA  

Publication
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)   Vol.J100-D   No.10   pp.897-901
Publication Date: 2017/10/01
Online ISSN: 1881-0225
Type of Manuscript: LETTER
Category: 
Keyword: 
image inpainting,  AR model,  robust estimation,  weighted joint optimization,  

Full Text(in Japanese): PDF(755.2KB)
>>Buy this Article


Summary: 
Conventional AR model based image inpainting algorithm do not perform well at non texture component in the image because these algorithms assume that an image is represented by an AR model. This letter formulates the image inpainting problem as joint optimization problem at modeling error and weighted TV norm.