Mathematical Analysis of DCN-Based Super-Resolution

C. Lee; J. Park; S. Woo; J. Kim; J. Yoon

doi:10.1109/ACCESS.2020.2992483

Mathematical Analysis of DCN-Based Super-Resolution

C. Lee, J. Park, S. Woo, J. Kim, J. Yoon

Department of Electrical and Electronic Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Although DCN-based super-resolution (DCN-SR) techniques have shown impressive performance, the working mechanism has not been completely understood and DCN-SR methods still produce some artefacts. In this paper, we analyze the working mechanisms of DCN-SR methods. We derive mathematical formulations of the DCN-SR methods and provide some experimental analyses, which show that the effective receptive fields of the DCN-SR methods are considerably smaller than the theoretical receptive fields. Based on the mathematical formulations, experiments were performed. The results indicates that current DCN-SR methods may have some fundamental problems and new types of DCN structures are needed for reliable super-resolution performance.

Original language	English
Article number	9086595
Pages (from-to)	90420-90429
Number of pages	10
Journal	IEEE Access
Volume	8
DOIs	https://doi.org/10.1109/ACCESS.2020.2992483
Publication status	Published - 2020

Bibliographical note

Funding Information:
This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education under Grant 2018R1D1A1B07050345.

All Science Journal Classification (ASJC) codes

Computer Science(all)
Materials Science(all)
Engineering(all)

Access to Document

10.1109/ACCESS.2020.2992483

Fingerprint Dive into the research topics of 'Mathematical Analysis of DCN-Based Super-Resolution'. Together they form a unique fingerprint.

Cite this

@article{3cf37ea22fef4ddd8ba6ce6c501ef9ed,

title = "Mathematical Analysis of DCN-Based Super-Resolution",

abstract = "Although DCN-based super-resolution (DCN-SR) techniques have shown impressive performance, the working mechanism has not been completely understood and DCN-SR methods still produce some artefacts. In this paper, we analyze the working mechanisms of DCN-SR methods. We derive mathematical formulations of the DCN-SR methods and provide some experimental analyses, which show that the effective receptive fields of the DCN-SR methods are considerably smaller than the theoretical receptive fields. Based on the mathematical formulations, experiments were performed. The results indicates that current DCN-SR methods may have some fundamental problems and new types of DCN structures are needed for reliable super-resolution performance.",

author = "C. Lee and J. Park and S. Woo and J. Kim and J. Yoon",

note = "Funding Information: This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education under Grant 2018R1D1A1B07050345.",

year = "2020",

doi = "10.1109/ACCESS.2020.2992483",

language = "English",

volume = "8",

pages = "90420--90429",

journal = "IEEE Access",

issn = "2169-3536",

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

}

TY - JOUR

T1 - Mathematical Analysis of DCN-Based Super-Resolution

AU - Lee, C.

AU - Park, J.

AU - Woo, S.

AU - Kim, J.

AU - Yoon, J.

N1 - Funding Information: This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education under Grant 2018R1D1A1B07050345.

PY - 2020

Y1 - 2020

N2 - Although DCN-based super-resolution (DCN-SR) techniques have shown impressive performance, the working mechanism has not been completely understood and DCN-SR methods still produce some artefacts. In this paper, we analyze the working mechanisms of DCN-SR methods. We derive mathematical formulations of the DCN-SR methods and provide some experimental analyses, which show that the effective receptive fields of the DCN-SR methods are considerably smaller than the theoretical receptive fields. Based on the mathematical formulations, experiments were performed. The results indicates that current DCN-SR methods may have some fundamental problems and new types of DCN structures are needed for reliable super-resolution performance.

AB - Although DCN-based super-resolution (DCN-SR) techniques have shown impressive performance, the working mechanism has not been completely understood and DCN-SR methods still produce some artefacts. In this paper, we analyze the working mechanisms of DCN-SR methods. We derive mathematical formulations of the DCN-SR methods and provide some experimental analyses, which show that the effective receptive fields of the DCN-SR methods are considerably smaller than the theoretical receptive fields. Based on the mathematical formulations, experiments were performed. The results indicates that current DCN-SR methods may have some fundamental problems and new types of DCN structures are needed for reliable super-resolution performance.

UR - http://www.scopus.com/inward/record.url?scp=85085570636&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85085570636&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2020.2992483

DO - 10.1109/ACCESS.2020.2992483

M3 - Article

AN - SCOPUS:85085570636

VL - 8

SP - 90420

EP - 90429

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 9086595

ER -