Multimodal Approach for Classifying Stress using Facial Emotion and Physiological Sensors

The focus is on understanding emotional stress by itself may enhance artificial intelligence agents involved in emotion detection and human computer interaction. These emotional responses are reflected into emotions and facial expressions. This research work proposes a study of Stress Classification...

Full description

Saved in:
Bibliographic Details
Published in2023 International Conference on Networking and Communications (ICNWC) pp. 1 - 5
Main Authors Abirami, M.S., Shringi, Umang, Mishra, Aditya
Format Conference Proceeding
LanguageEnglish
Published IEEE 05.04.2023
Subjects
Anxiety disorders
Approximation algorithms
Convolutional Neural network
Deep learning
Image sensors
Machine learning
Physiological Sensors
Sensor phenomena and characterization
Temperature distribution
Temperature sensors
Transfer learning
Online AccessGet full text

Cover

More Information
Summary:The focus is on understanding emotional stress by itself may enhance artificial intelligence agents involved in emotion detection and human computer interaction. These emotional responses are reflected into emotions and facial expressions. This research work proposes a study of Stress Classification using both facial expression and Physiological Sensors. For getting facial data, transfer learning is used with fine-tuning to extract features from facial images. In transfer learning different Deep Learning architectures like VGG-19[13], ResNet are used. From the sensors data on four features are collected that is age, gender, body temperature and heartbeat and accordingly choose the architecture for doing stress classification. Finally, both these models are integrated for getting better results.
DOI:10.1109/ICNWC57852.2023.10127519