Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution

Keon Hee Ahn; Jun Hyuk Kim; Jun Ho Choi; Jong Seok Lee

doi:10.1109/ICCE-Asia49877.2020.9276925

Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution

Keon Hee Ahn, Jun Hyuk Kim, Jun Ho Choi, Jong Seok Lee

School of Integrated Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Single image super-resolution (SISR) has developed fast for recent years. Most of the SISR models are trained and evaluated with simulated data where low-resolution (LR) images are generated from high-resolution (HR) images using pre-defined degradation. In contrast, real-world image super-resolution (RealSR) is more challenging since the process of obtaining LR images is formulated by complex degradation. To solve this problem, we propose the multi-scale adaptive real image super-resolution (MARS). Our model extracts complex features in the image and uses them for upscaling adaptively. Experimental results show that the proposed method can improve the quality of the super-resolved images in RealSR.

Original language	English
Title of host publication	2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728161648
DOIs	https://doi.org/10.1109/ICCE-Asia49877.2020.9276925
Publication status	Published - 2020 Nov 1
Event	2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020 - Seoul, Korea, Republic of Duration: 2020 Nov 1 → 2020 Nov 3

Publication series

Name	2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020

Conference

Conference	2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020
Country/Territory	Korea, Republic of
City	Seoul
Period	20/11/1 → 20/11/3

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

All Science Journal Classification (ASJC) codes

Computer Science Applications
Electrical and Electronic Engineering
Media Technology
Instrumentation

Access to Document

10.1109/ICCE-Asia49877.2020.9276925

Cite this

Ahn, K. H., Kim, J. H., Choi, J. H., & Lee, J. S. (2020). Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution. In 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020 Article 9276925 (2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCE-Asia49877.2020.9276925

Ahn, Keon Hee ; Kim, Jun Hyuk ; Choi, Jun Ho et al. / Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution. 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020. Institute of Electrical and Electronics Engineers Inc., 2020. (2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020).

@inproceedings{59d7b98b535242ed811c8e0b84c79169,

title = "Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution",

abstract = "Single image super-resolution (SISR) has developed fast for recent years. Most of the SISR models are trained and evaluated with simulated data where low-resolution (LR) images are generated from high-resolution (HR) images using pre-defined degradation. In contrast, real-world image super-resolution (RealSR) is more challenging since the process of obtaining LR images is formulated by complex degradation. To solve this problem, we propose the multi-scale adaptive real image super-resolution (MARS). Our model extracts complex features in the image and uses them for upscaling adaptively. Experimental results show that the proposed method can improve the quality of the super-resolved images in RealSR.",

keywords = "Single image super-resolution, attention mechanism, real-world image super-resolution, total variation",

author = "Ahn, {Keon Hee} and Kim, {Jun Hyuk} and Choi, {Jun Ho} and Lee, {Jong Seok}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020 ; Conference date: 01-11-2020 Through 03-11-2020",

year = "2020",

month = nov,

day = "1",

doi = "10.1109/ICCE-Asia49877.2020.9276925",

language = "English",

series = "2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020",

address = "United States",

}

Ahn, KH, Kim, JH, Choi, JH & Lee, JS 2020, Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution. in 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020., 9276925, 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020, Institute of Electrical and Electronics Engineers Inc., 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020, Seoul, Korea, Republic of, 20/11/1. https://doi.org/10.1109/ICCE-Asia49877.2020.9276925

Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution. / Ahn, Keon Hee; Kim, Jun Hyuk; Choi, Jun Ho et al.
2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020. Institute of Electrical and Electronics Engineers Inc., 2020. 9276925 (2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution

AU - Ahn, Keon Hee

AU - Kim, Jun Hyuk

AU - Choi, Jun Ho

AU - Lee, Jong Seok

PY - 2020/11/1

Y1 - 2020/11/1

N2 - Single image super-resolution (SISR) has developed fast for recent years. Most of the SISR models are trained and evaluated with simulated data where low-resolution (LR) images are generated from high-resolution (HR) images using pre-defined degradation. In contrast, real-world image super-resolution (RealSR) is more challenging since the process of obtaining LR images is formulated by complex degradation. To solve this problem, we propose the multi-scale adaptive real image super-resolution (MARS). Our model extracts complex features in the image and uses them for upscaling adaptively. Experimental results show that the proposed method can improve the quality of the super-resolved images in RealSR.

AB - Single image super-resolution (SISR) has developed fast for recent years. Most of the SISR models are trained and evaluated with simulated data where low-resolution (LR) images are generated from high-resolution (HR) images using pre-defined degradation. In contrast, real-world image super-resolution (RealSR) is more challenging since the process of obtaining LR images is formulated by complex degradation. To solve this problem, we propose the multi-scale adaptive real image super-resolution (MARS). Our model extracts complex features in the image and uses them for upscaling adaptively. Experimental results show that the proposed method can improve the quality of the super-resolved images in RealSR.

KW - Single image super-resolution

KW - attention mechanism

KW - real-world image super-resolution

KW - total variation

UR - http://www.scopus.com/inward/record.url?scp=85098875784&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85098875784&partnerID=8YFLogxK

U2 - 10.1109/ICCE-Asia49877.2020.9276925

DO - 10.1109/ICCE-Asia49877.2020.9276925

M3 - Conference contribution

AN - SCOPUS:85098875784

T3 - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020

BT - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020

Y2 - 1 November 2020 through 3 November 2020

ER -

Ahn KH, Kim JH, Choi JH, Lee JS. Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution. In 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020. Institute of Electrical and Electronics Engineers Inc. 2020. 9276925. (2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020). doi: 10.1109/ICCE-Asia49877.2020.9276925

Multi-scale Adaptive Residual Network Using Total Variation for Real Image Super-Resolution

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this