Single image super-resolution with limited number of filters

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a single image super-resolution with limited number of filters based on RAISR. RAISR is well known as rapid and accurate super-resolution method which utilizes 864 filters for upscaling. This super-resolution idea utilizes the filter learned with sucient training set. To get low cost of calculation and comparable image quality with other highly accurate super-resolution methods, the patch of input image is classified into classes by simple hash calculation. Then, the high quality version of this patch is generated by applying the filter to low resolution patches. In our method, only 18 filters can make high resolution images by using simple geometric conversion and rotation conversion while keeping the accuracy and runtime of RAISR.

Original languageEnglish
Title of host publication2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages36-40
Number of pages5
ISBN (Electronic)9781728112954
DOIs
Publication statusPublished - 2019 Feb 20
Event2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Anaheim, United States
Duration: 2018 Nov 262018 Nov 29

Publication series

Name2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

Conference

Conference2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018
CountryUnited States
CityAnaheim
Period18/11/2618/11/29

Fingerprint

Image resolution
Image quality
Costs

Keywords

  • Filter learning
  • Image enhancement
  • Super-resolution

ASJC Scopus subject areas

  • Information Systems
  • Signal Processing

Cite this

Nakahara, Y., Yamaguchi, T., & Ikehara, M. (2019). Single image super-resolution with limited number of filters. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings (pp. 36-40). [8646455] (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2018.8646455

Single image super-resolution with limited number of filters. / Nakahara, Yusuke; Yamaguchi, Takuro; Ikehara, Masaaki.

2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 36-40 8646455 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nakahara, Y, Yamaguchi, T & Ikehara, M 2019, Single image super-resolution with limited number of filters. in 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings., 8646455, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 36-40, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018, Anaheim, United States, 18/11/26. https://doi.org/10.1109/GlobalSIP.2018.8646455
Nakahara Y, Yamaguchi T, Ikehara M. Single image super-resolution with limited number of filters. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 36-40. 8646455. (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). https://doi.org/10.1109/GlobalSIP.2018.8646455
Nakahara, Yusuke ; Yamaguchi, Takuro ; Ikehara, Masaaki. / Single image super-resolution with limited number of filters. 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 36-40 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).
@inproceedings{fbccf8d0e59d4c89a6ebbbf0e168a659,
title = "Single image super-resolution with limited number of filters",
abstract = "In this paper, we propose a single image super-resolution with limited number of filters based on RAISR. RAISR is well known as rapid and accurate super-resolution method which utilizes 864 filters for upscaling. This super-resolution idea utilizes the filter learned with sucient training set. To get low cost of calculation and comparable image quality with other highly accurate super-resolution methods, the patch of input image is classified into classes by simple hash calculation. Then, the high quality version of this patch is generated by applying the filter to low resolution patches. In our method, only 18 filters can make high resolution images by using simple geometric conversion and rotation conversion while keeping the accuracy and runtime of RAISR.",
keywords = "Filter learning, Image enhancement, Super-resolution",
author = "Yusuke Nakahara and Takuro Yamaguchi and Masaaki Ikehara",
year = "2019",
month = "2",
day = "20",
doi = "10.1109/GlobalSIP.2018.8646455",
language = "English",
series = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "36--40",
booktitle = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",

}

TY - GEN

T1 - Single image super-resolution with limited number of filters

AU - Nakahara, Yusuke

AU - Yamaguchi, Takuro

AU - Ikehara, Masaaki

PY - 2019/2/20

Y1 - 2019/2/20

N2 - In this paper, we propose a single image super-resolution with limited number of filters based on RAISR. RAISR is well known as rapid and accurate super-resolution method which utilizes 864 filters for upscaling. This super-resolution idea utilizes the filter learned with sucient training set. To get low cost of calculation and comparable image quality with other highly accurate super-resolution methods, the patch of input image is classified into classes by simple hash calculation. Then, the high quality version of this patch is generated by applying the filter to low resolution patches. In our method, only 18 filters can make high resolution images by using simple geometric conversion and rotation conversion while keeping the accuracy and runtime of RAISR.

AB - In this paper, we propose a single image super-resolution with limited number of filters based on RAISR. RAISR is well known as rapid and accurate super-resolution method which utilizes 864 filters for upscaling. This super-resolution idea utilizes the filter learned with sucient training set. To get low cost of calculation and comparable image quality with other highly accurate super-resolution methods, the patch of input image is classified into classes by simple hash calculation. Then, the high quality version of this patch is generated by applying the filter to low resolution patches. In our method, only 18 filters can make high resolution images by using simple geometric conversion and rotation conversion while keeping the accuracy and runtime of RAISR.

KW - Filter learning

KW - Image enhancement

KW - Super-resolution

UR - http://www.scopus.com/inward/record.url?scp=85063111809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063111809&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2018.8646455

DO - 10.1109/GlobalSIP.2018.8646455

M3 - Conference contribution

AN - SCOPUS:85063111809

T3 - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

SP - 36

EP - 40

BT - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -