14 May 2018 High performance image completion using sparsity based algorithms
Author Affiliations +
Abstract
There are many applications that have corrupted or missing pixels in images. Here, we present sparsity based image completion algorithms that can achieve high performance in image reconstruction. Through extensive experiments using various types of images, it was demonstrated that our algorithms can deal with extremely high missing rates (up to 99.9%) and relatively large missing blocks.
Conference Presentation
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jin Zhou, Jin Zhou, Chiman Kwan, Chiman Kwan, } "High performance image completion using sparsity based algorithms", Proc. SPIE 10669, Computational Imaging III, 106690I (14 May 2018); doi: 10.1117/12.2303661; https://doi.org/10.1117/12.2303661
PROCEEDINGS
17 PAGES + PRESENTATION

SHARE
Back to Top