Home // ICCGI 2016, The Eleventh International Multi-Conference on Computing in the Global Information Technology // View article

Subjective Assessment of Super Resolution for Remastering on 4K TVs

Authors:
Hiroki Shoji
Seiichi Gohshi

Keywords: Learning-Based Super-Resolution; Non-Linear Signal Processing; 4K TV; Subjective Assessment; Performance Verification.

Abstract:
Super-resolution (SR) is a technology to create high-definition images. According to television (TV) manufacturer's advertisements, TVs sold recently in Japan have SR functions. In Japan, when such TVs are sold, SR is aggressively advertised on a large scale; however, in countries other than Japan, SR is not mentioned in similar TV manufacturer's advertisements. In previous research, real-time processing to generate SR images has been found to be difficult. It is necessary to verify whether SR advertised by TV manufacturers exhibits its original performance in a TV that requires a real-time processing. However, an objective assessment of SR on TVs cannot be conducted because images processed in TVs cannot be extracted. Therefore, in our previous work, a subjective assessment of Learning-Based Super-Resolution (LBSR) was conducted, and it was shown that the subjective assessment is effective in performance verification of SR on a TV. Moreover, we conducted a subjective assessment of LBSR and Non-Linear Signal Processing (NLSP) using a 4K TV to evaluate the image quality of each SR image produced via up-conversion from HD video to 4K. In this study, the image quality of each SR when remastering 4K video to 4K is evaluated by the subjective assessment. Furthermore, the performance results of both LBSR and NLSP for remastering on a 4K TV are reported.

Pages: 10 to 15

Copyright: Copyright (c) IARIA, 2016

Publication date: November 13, 2016

Published in: conference

ISSN: 2308-4529

ISBN: 978-1-61208-513-5

Location: Barcelona, Spain

Dates: from November 13, 2016 to November 17, 2016