Facial expression recognition on partial facial sections

Ryan Melaugh; Nazmul Siddique; Sonya Coleman; Pratheepan Yogarajah

doi:10.1109/ISPA.2019.8868630

Facial expression recognition on partial facial sections

Ryan Melaugh, Nazmul Siddique, Sonya Coleman, Pratheepan Yogarajah

Intelligent Systems Research Centre

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

12 Citations (Scopus)

179 Downloads (Pure)

Abstract

Research by psychologists have shown that subjects had a preference for a side of a face when it was expressing emotions. This paper seeks to find what accuracies can be attained when only a segment of the face is considered. We show that using one side of the face only reduces accuracy by 0.34% but at half the computationally time required. Various other sections of the face are evaluated for similar performance. We demonstrate that using smaller portions of the face have an expected computation reduction but dont suffer the same degree of accuracy loss. For evaluating we train with a Convolutional Neural Network. To test what portions of a facial image are useful, the full face, half face, eyes, single eye, mouth and half of the mouth are chosen. These images come from the JAFFE, CK+ and KDEF datasets.

Original language	English
Title of host publication	ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis
Editors	Sven Loncaric, Robert Bregovic, Marco Carli, Marko Subasic
Publisher	IEEE Computer Society
Pages	193-197
Number of pages	5
ISBN (Electronic)	9781728131405
DOIs	https://doi.org/10.1109/ISPA.2019.8868630
Publication status	Published (in print/issue) - 17 Oct 2019
Event	11th International Symposium on Image and Signal Processing and Analysis, ISPA 2019 - Dubrovnik, Croatia Duration: 23 Sept 2019 → 25 Sept 2019

Publication series

Name	International Symposium on Image and Signal Processing and Analysis, ISPA
Volume	2019-September
ISSN (Print)	1845-5921
ISSN (Electronic)	1849-2266

Conference

Conference	11th International Symposium on Image and Signal Processing and Analysis, ISPA 2019
Country/Territory	Croatia
City	Dubrovnik
Period	23/09/19 → 25/09/19

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.

Keywords

CNN
Facial Expression Recognition
Hemisphere differences
Image Processing
Neural Network
Occlusion

Access to Document

10.1109/ISPA.2019.8868630

accepted version SColemanAuthor Accepted Manuscript, 320 KB

Cite this

Melaugh, R., Siddique, N., Coleman, S., & Yogarajah, P. (2019). Facial expression recognition on partial facial sections. In S. Loncaric, R. Bregovic, M. Carli, & M. Subasic (Eds.), ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis (pp. 193-197). Article 8868630 (International Symposium on Image and Signal Processing and Analysis, ISPA; Vol. 2019-September). IEEE Computer Society. https://doi.org/10.1109/ISPA.2019.8868630

Melaugh, Ryan ; Siddique, Nazmul ; Coleman, Sonya et al. / Facial expression recognition on partial facial sections. ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis. editor / Sven Loncaric ; Robert Bregovic ; Marco Carli ; Marko Subasic. IEEE Computer Society, 2019. pp. 193-197 (International Symposium on Image and Signal Processing and Analysis, ISPA).

@inproceedings{6b6c1fa49cbf48e5986421ce1dffd4d3,

title = "Facial expression recognition on partial facial sections",

abstract = "Research by psychologists have shown that subjects had a preference for a side of a face when it was expressing emotions. This paper seeks to find what accuracies can be attained when only a segment of the face is considered. We show that using one side of the face only reduces accuracy by 0.34% but at half the computationally time required. Various other sections of the face are evaluated for similar performance. We demonstrate that using smaller portions of the face have an expected computation reduction but dont suffer the same degree of accuracy loss. For evaluating we train with a Convolutional Neural Network. To test what portions of a facial image are useful, the full face, half face, eyes, single eye, mouth and half of the mouth are chosen. These images come from the JAFFE, CK+ and KDEF datasets.",

keywords = "CNN, Facial Expression Recognition, Hemisphere differences, Image Processing, Neural Network, Occlusion",

author = "Ryan Melaugh and Nazmul Siddique and Sonya Coleman and Pratheepan Yogarajah",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; 11th International Symposium on Image and Signal Processing and Analysis, ISPA 2019 ; Conference date: 23-09-2019 Through 25-09-2019",

year = "2019",

month = oct,

day = "17",

doi = "10.1109/ISPA.2019.8868630",

language = "English",

series = "International Symposium on Image and Signal Processing and Analysis, ISPA",

publisher = "IEEE Computer Society",

pages = "193--197",

editor = "Sven Loncaric and Robert Bregovic and Marco Carli and Marko Subasic",

booktitle = "ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis",

address = "United States",

}

Melaugh, R, Siddique, N , Coleman, S & Yogarajah, P 2019, Facial expression recognition on partial facial sections. in S Loncaric, R Bregovic, M Carli & M Subasic (eds), ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis., 8868630, International Symposium on Image and Signal Processing and Analysis, ISPA, vol. 2019-September, IEEE Computer Society, pp. 193-197, 11th International Symposium on Image and Signal Processing and Analysis, ISPA 2019, Dubrovnik, Croatia, 23/09/19. https://doi.org/10.1109/ISPA.2019.8868630

Facial expression recognition on partial facial sections. / Melaugh, Ryan; Siddique, Nazmul ; Coleman, Sonya et al.
ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis. ed. / Sven Loncaric; Robert Bregovic; Marco Carli; Marko Subasic. IEEE Computer Society, 2019. p. 193-197 8868630 (International Symposium on Image and Signal Processing and Analysis, ISPA; Vol. 2019-September).

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

TY - GEN

T1 - Facial expression recognition on partial facial sections

AU - Melaugh, Ryan

AU - Siddique, Nazmul

AU - Coleman, Sonya

AU - Yogarajah, Pratheepan

PY - 2019/10/17

Y1 - 2019/10/17

N2 - Research by psychologists have shown that subjects had a preference for a side of a face when it was expressing emotions. This paper seeks to find what accuracies can be attained when only a segment of the face is considered. We show that using one side of the face only reduces accuracy by 0.34% but at half the computationally time required. Various other sections of the face are evaluated for similar performance. We demonstrate that using smaller portions of the face have an expected computation reduction but dont suffer the same degree of accuracy loss. For evaluating we train with a Convolutional Neural Network. To test what portions of a facial image are useful, the full face, half face, eyes, single eye, mouth and half of the mouth are chosen. These images come from the JAFFE, CK+ and KDEF datasets.

AB - Research by psychologists have shown that subjects had a preference for a side of a face when it was expressing emotions. This paper seeks to find what accuracies can be attained when only a segment of the face is considered. We show that using one side of the face only reduces accuracy by 0.34% but at half the computationally time required. Various other sections of the face are evaluated for similar performance. We demonstrate that using smaller portions of the face have an expected computation reduction but dont suffer the same degree of accuracy loss. For evaluating we train with a Convolutional Neural Network. To test what portions of a facial image are useful, the full face, half face, eyes, single eye, mouth and half of the mouth are chosen. These images come from the JAFFE, CK+ and KDEF datasets.

KW - CNN

KW - Facial Expression Recognition

KW - Hemisphere differences

KW - Image Processing

KW - Neural Network

KW - Occlusion

UR - http://www.scopus.com/inward/record.url?scp=85074418074&partnerID=8YFLogxK

U2 - 10.1109/ISPA.2019.8868630

DO - 10.1109/ISPA.2019.8868630

M3 - Conference contribution

AN - SCOPUS:85074418074

T3 - International Symposium on Image and Signal Processing and Analysis, ISPA

SP - 193

EP - 197

BT - ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis

A2 - Loncaric, Sven

A2 - Bregovic, Robert

A2 - Carli, Marco

A2 - Subasic, Marko

PB - IEEE Computer Society

T2 - 11th International Symposium on Image and Signal Processing and Analysis, ISPA 2019

Y2 - 23 September 2019 through 25 September 2019

ER -

Melaugh R, Siddique N , Coleman S , Yogarajah P. Facial expression recognition on partial facial sections. In Loncaric S, Bregovic R, Carli M, Subasic M, editors, ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis. IEEE Computer Society. 2019. p. 193-197. 8868630. (International Symposium on Image and Signal Processing and Analysis, ISPA). doi: 10.1109/ISPA.2019.8868630

Facial expression recognition on partial facial sections

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this