VIP Lab’s paper accepted to CVPR 2017

Activities March 3, 2017

UNIST ECE student, Byeong-Ju Han and Prof. Jae-Young Sim’s paper has been accepted to CVPR 2017.


Authors: Byeong-Ju Han and Jae-Young Sim (corresponding author)

Title: Reflection removal using low-rank matrix completion

Conference: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)


The images taken through glass often capture a target transmitted scene as well as undesired reflected scenes. In this paper, the authors proposed a novel algorithm to recover desired target scene and remove reflection artifacts from glass images automatically.

CVPR is one of the most prestige international conferences in the field of computer vision, and it is ranked in the top 50 of all publications in Google Scholar.