Comparative analysis of facial expression recognition system for age, gender, and emotions

C. Sheeba Joice; C. Jenisha; S. Hitha Shanthini; R. Vedhapriyavadhana

doi:10.1201/9781003587538

Comparative analysis of facial expression recognition system for age, gender, and emotions

C. Sheeba Joice, C. Jenisha, S. Hitha Shanthini, R. Vedhapriyavadhana

School of Computing, Engineering and Physical Sciences

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

Abstract

Facial expression recognition systems have advanced quickly, allowing for accurate real-time detection of age, gender, and emotion. Facial features are retrieved effectively for exact identification using deep learning models such as MobileNetV2, ResNet, and DenseNet, as well as approaches like the Haar cascade classifier. The use of datasets such as FER 2013 promotes robust model training, while preprocessing procedures provide peak performance. Models such as the Caffe model achieve great accuracy in detecting age and gender in real time through transfer learning and fine-tuning. These improvements highlight the potential of facial recognition systems in a variety of applications, including enhanced security, healthcare, and human-computer interaction.

Original language	English
Title of host publication	Pedagogical Revelations and Emerging Trends
Editors	C. Sheeba Joice, M. Selvi
Place of Publication	London
Publisher	CRC Press
Chapter	50
ISBN (Electronic)	9781003587538
ISBN (Print)	9781032960012, 9781032960029
DOIs	https://doi.org/10.1201/9781003587538
Publication status	Published - 27 Jan 2025

Keywords

facial expression recognition (FER)
deep learning
emotions
age

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1201/9781003587538

https://www.routledge.com/Pedagogical-Revelations-and-Emerging-Trends/Joice-Selvi/p/book/9781032960029?_gl=1*lfmany*_gcl_au*MTc0NDUwMTYzNy4xNzQzMDkwNDU0*_ga*MTM1OTIzOTU2My4xNzQzMDkwNDU2*_ga_0HYE8YG0M6*czE3NDk1NjQwMzIkbzckZzEkdDE3NDk1NjQwODAkajEyJGwwJGgw

Embargoed Document

2024 09 15 Joice et al Expression accepted - for Pure
Accepted author manuscript, 172 KB
Licence: CC BY-NC-ND
Embargo ends: 27/01/26

Cite this

@inbook{47fab3e1911042b09c585e70443006fd,

title = "Comparative analysis of facial expression recognition system for age, gender, and emotions",

abstract = "Facial expression recognition systems have advanced quickly, allowing for accurate real-time detection of age, gender, and emotion. Facial features are retrieved effectively for exact identification using deep learning models such as MobileNetV2, ResNet, and DenseNet, as well as approaches like the Haar cascade classifier. The use of datasets such as FER 2013 promotes robust model training, while preprocessing procedures provide peak performance. Models such as the Caffe model achieve great accuracy in detecting age and gender in real time through transfer learning and fine-tuning. These improvements highlight the potential of facial recognition systems in a variety of applications, including enhanced security, healthcare, and human-computer interaction.",

keywords = "facial expression recognition (FER), deep learning, emotions, age",

author = "{Sheeba Joice}, C. and C. Jenisha and {Hitha Shanthini}, S. and R. Vedhapriyavadhana",

year = "2025",

month = jan,

day = "27",

doi = "10.1201/9781003587538",

language = "English",

isbn = "9781032960012",

editor = "{Sheeba Joice}, C. and M. Selvi",

booktitle = "Pedagogical Revelations and Emerging Trends",

publisher = "CRC Press",

}

TY - CHAP

T1 - Comparative analysis of facial expression recognition system for age, gender, and emotions

AU - Sheeba Joice, C.

AU - Jenisha, C.

AU - Hitha Shanthini, S.

AU - Vedhapriyavadhana, R.

PY - 2025/1/27

Y1 - 2025/1/27

N2 - Facial expression recognition systems have advanced quickly, allowing for accurate real-time detection of age, gender, and emotion. Facial features are retrieved effectively for exact identification using deep learning models such as MobileNetV2, ResNet, and DenseNet, as well as approaches like the Haar cascade classifier. The use of datasets such as FER 2013 promotes robust model training, while preprocessing procedures provide peak performance. Models such as the Caffe model achieve great accuracy in detecting age and gender in real time through transfer learning and fine-tuning. These improvements highlight the potential of facial recognition systems in a variety of applications, including enhanced security, healthcare, and human-computer interaction.

AB - Facial expression recognition systems have advanced quickly, allowing for accurate real-time detection of age, gender, and emotion. Facial features are retrieved effectively for exact identification using deep learning models such as MobileNetV2, ResNet, and DenseNet, as well as approaches like the Haar cascade classifier. The use of datasets such as FER 2013 promotes robust model training, while preprocessing procedures provide peak performance. Models such as the Caffe model achieve great accuracy in detecting age and gender in real time through transfer learning and fine-tuning. These improvements highlight the potential of facial recognition systems in a variety of applications, including enhanced security, healthcare, and human-computer interaction.

KW - facial expression recognition (FER)

KW - deep learning

KW - emotions

KW - age

U2 - 10.1201/9781003587538

DO - 10.1201/9781003587538

M3 - Chapter

SN - 9781032960012

SN - 9781032960029

BT - Pedagogical Revelations and Emerging Trends

A2 - Sheeba Joice, C.

A2 - Selvi, M.

PB - CRC Press

CY - London

ER -

Comparative analysis of facial expression recognition system for age, gender, and emotions

Abstract

Keywords

UN SDGs

Access to Document

Embargoed Document

Fingerprint

Cite this