TMSR: Tiny Multi-path CNNs for Super Resolution

Chia Hung Liu; Tzu Hsin Hsieh; Kuan Yu Huang; Pei Yin Chen

doi:10.1109/ECICE59523.2023.10383101

TMSR: Tiny Multi-path CNNs for Super Resolution

Chia Hung Liu
, Tzu Hsin Hsieh
, Kuan Yu Huang
, Pei Yin Chen

Department of Computer Science and Information Engineering

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

Abstract

We proposed a tiny multi-path CNN-based Super-Resolution (SR) method (TMSR) referring to tiny CNN-based SR methods under 5k parameters. The main contribution of the proposed method is the improved multi-path learning. The experimental results showed that TMSR obtained competitive image quality (peak-signal-to-noise ratio and Structural Similarity Index Measure) compared to the related works under 5k parameters.

Original language	English
Title of host publication	2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023
Editors	Teen-Hang Meen
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	829-833
Number of pages	5
ISBN (Electronic)	9798350314694
DOIs	https://doi.org/10.1109/ECICE59523.2023.10383101
Publication status	Published - 2023
Event	5th IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2023 - Yunlin, Taiwan Duration: 2023 Oct 27 → 2023 Oct 29

Publication series

Name	2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023

Conference

Conference	5th IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2023
Country/Territory	Taiwan
City	Yunlin
Period	23-10-27 → 23-10-29

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications
Computer Vision and Pattern Recognition
Hardware and Architecture
Human-Computer Interaction
Media Technology

Access to Document

10.1109/ECICE59523.2023.10383101

Cite this

Liu, C. H., Hsieh, T. H., Huang, K. Y., & Chen, P. Y. (2023). TMSR: Tiny Multi-path CNNs for Super Resolution. In T.-H. Meen (Ed.), 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023 (pp. 829-833). (2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECICE59523.2023.10383101

Liu, Chia Hung ; Hsieh, Tzu Hsin ; Huang, Kuan Yu et al. / TMSR : Tiny Multi-path CNNs for Super Resolution. 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023. editor / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 829-833 (2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023).

@inproceedings{6a0f64024cba48a9a05078ed006a3232,

title = "TMSR: Tiny Multi-path CNNs for Super Resolution",

abstract = "We proposed a tiny multi-path CNN-based Super-Resolution (SR) method (TMSR) referring to tiny CNN-based SR methods under 5k parameters. The main contribution of the proposed method is the improved multi-path learning. The experimental results showed that TMSR obtained competitive image quality (peak-signal-to-noise ratio and Structural Similarity Index Measure) compared to the related works under 5k parameters.",

author = "Liu, \{Chia Hung\} and Hsieh, \{Tzu Hsin\} and Huang, \{Kuan Yu\} and Chen, \{Pei Yin\}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 5th IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2023 ; Conference date: 27-10-2023 Through 29-10-2023",

year = "2023",

doi = "10.1109/ECICE59523.2023.10383101",

language = "English",

series = "2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023",

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

pages = "829--833",

editor = "Teen-Hang Meen",

booktitle = "2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023",

address = "United States",

}

Liu, CH, Hsieh, TH, Huang, KY & Chen, PY 2023, TMSR: Tiny Multi-path CNNs for Super Resolution. in T-H Meen (ed.), 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023. 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023, Institute of Electrical and Electronics Engineers Inc., pp. 829-833, 5th IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2023, Yunlin, Taiwan, 23-10-27. https://doi.org/10.1109/ECICE59523.2023.10383101

TMSR: Tiny Multi-path CNNs for Super Resolution. / Liu, Chia Hung; Hsieh, Tzu Hsin; Huang, Kuan Yu et al.
2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023. ed. / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2023. p. 829-833 (2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023).

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

TY - GEN

T1 - TMSR

T2 - 5th IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2023

AU - Liu, Chia Hung

AU - Hsieh, Tzu Hsin

AU - Huang, Kuan Yu

AU - Chen, Pei Yin

PY - 2023

Y1 - 2023

N2 - We proposed a tiny multi-path CNN-based Super-Resolution (SR) method (TMSR) referring to tiny CNN-based SR methods under 5k parameters. The main contribution of the proposed method is the improved multi-path learning. The experimental results showed that TMSR obtained competitive image quality (peak-signal-to-noise ratio and Structural Similarity Index Measure) compared to the related works under 5k parameters.

AB - We proposed a tiny multi-path CNN-based Super-Resolution (SR) method (TMSR) referring to tiny CNN-based SR methods under 5k parameters. The main contribution of the proposed method is the improved multi-path learning. The experimental results showed that TMSR obtained competitive image quality (peak-signal-to-noise ratio and Structural Similarity Index Measure) compared to the related works under 5k parameters.

UR - https://www.scopus.com/pages/publications/85184082020

UR - https://www.scopus.com/pages/publications/85184082020#tab=citedBy

U2 - 10.1109/ECICE59523.2023.10383101

DO - 10.1109/ECICE59523.2023.10383101

M3 - Conference contribution

AN - SCOPUS:85184082020

T3 - 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023

SP - 829

EP - 833

BT - 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023

A2 - Meen, Teen-Hang

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 27 October 2023 through 29 October 2023

ER -

Liu CH, Hsieh TH, Huang KY, Chen PY. TMSR: Tiny Multi-path CNNs for Super Resolution. In Meen TH, editor, 2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 829-833. (2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering, ECICE 2023). doi: 10.1109/ECICE59523.2023.10383101

TMSR: Tiny Multi-path CNNs for Super Resolution

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this