Application and Analysis of the VGG16 Model in Facial Emotion Recognition

Yitong Bai

2023

Abstract

This paper introduces a facial emotion analysis model developed through the utilization of the deep convolutional neural network structure known as Visual Geometry Group 16 (VGG16). exploring its significance and effectiveness in the field of psychotherapy. The research employs the Facial Expression Recognition 2013 (FER2013) dataset, consisting of 35,887 facial images covering various Categories of emotional states including anger, disgust, fear, happiness, sadness, surprise, and neutrality. VGG16 functions as a feature extraction tool, employing the derived multi-level features for emotion classification through a Multilayer Perceptron (MLP) classifier. Additionally, VGG16 is employed as an end-to-end sentiment classifier with structural and parameter optimization, incorporating techniques such as data augmentation and model fusion to enhance performance and stability. By applying the model in the domain of psychotherapy, its responsiveness and relevance in recognizing and regulating emotions associated with different psychological disorders are explored. Empirical study results demonstrate that the proposed facial emotion analysis method significantly improves emotion recognition accuracy and robustness. This research holds paramount importance in advancing the fields of human-computer interaction, mental health and education.

Paper Citation

in Harvard Style

Bai Y. (2023). Application and Analysis of the VGG16 Model in Facial Emotion Recognition. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 422-426. DOI: 10.5220/0012799800003885

in Bibtex Style

@conference{daml23,
author={Yitong Bai},
title={Application and Analysis of the VGG16 Model in Facial Emotion Recognition},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={422-426},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012799800003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Application and Analysis of the VGG16 Model in Facial Emotion Recognition
SN - 978-989-758-705-4
AU - Bai Y.
PY - 2023
SP - 422
EP - 426
DO - 10.5220/0012799800003885
PB - SciTePress