Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability

Independent, discrete models like Paul Ekman’s six basic emotions model are widely used in affective state assessment (ASA) and facial expression classification. However, the continuous and dynamic nature of human expressions often needs to be considered for accurately assessing facial expressio...

Full description

Bibliographic Details
Main Authors: Vice, Jordan, Khan, Masood, Tan, Tele, Yanushkevich, Svetlana
Other Authors: Papadopoulos, George
Format: Conference Paper
Language:English
Published: ieee.org 2022
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/88713
_version_ 1848765070701494272
author Vice, Jordan
Khan, Masood
Tan, Tele
Yanushkevich, Svetlana
author2 Papadopoulos, George
author_facet Papadopoulos, George
Vice, Jordan
Khan, Masood
Tan, Tele
Yanushkevich, Svetlana
author_sort Vice, Jordan
building Curtin Institutional Repository
collection Online Access
description Independent, discrete models like Paul Ekman’s six basic emotions model are widely used in affective state assessment (ASA) and facial expression classification. However, the continuous and dynamic nature of human expressions often needs to be considered for accurately assessing facial expressions of affective states. This paper investigates how mutual information-carrying continuous models can be extracted and used in continuous and dynamic facial expression classification systems for improving the efficacy and reliability of ASA systems. A novel, hybrid learning model that projects continuous data onto a multidimensional hyperplane is proposed. Through cosine similarity-based clustering (unsupervised) and classification (supervised) processes, our hybrid approach allows us to transform seven, discrete facial expression models into twenty-one facial expression models that include micro-expressions. The proposed continuous, dynamic classifier was able to achieve greater than 73% accuracy when experimented with Random Forest, Support Vector Machine (SVM) and Neural Network classification architectures. The presented system was validated using the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) and the extended Cohn-Kanade (CK+) dataset.
first_indexed 2025-11-14T11:29:24Z
format Conference Paper
id curtin-20.500.11937-88713
institution Curtin University Malaysia
institution_category Local University
language English
last_indexed 2025-11-14T11:29:24Z
publishDate 2022
publisher ieee.org
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-887132022-06-21T02:41:23Z Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability Vice, Jordan Khan, Masood Tan, Tele Yanushkevich, Svetlana Papadopoulos, George Angelov, Plamen 4601 - Applied computing 4602 - Artificial intelligence 4603 - Computer vision and multimedia computation 4611 - Machine learning 0915 - Interdisciplinary Engineering Independent, discrete models like Paul Ekman’s six basic emotions model are widely used in affective state assessment (ASA) and facial expression classification. However, the continuous and dynamic nature of human expressions often needs to be considered for accurately assessing facial expressions of affective states. This paper investigates how mutual information-carrying continuous models can be extracted and used in continuous and dynamic facial expression classification systems for improving the efficacy and reliability of ASA systems. A novel, hybrid learning model that projects continuous data onto a multidimensional hyperplane is proposed. Through cosine similarity-based clustering (unsupervised) and classification (supervised) processes, our hybrid approach allows us to transform seven, discrete facial expression models into twenty-one facial expression models that include micro-expressions. The proposed continuous, dynamic classifier was able to achieve greater than 73% accuracy when experimented with Random Forest, Support Vector Machine (SVM) and Neural Network classification architectures. The presented system was validated using the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) and the extended Cohn-Kanade (CK+) dataset. 2022 Conference Paper http://hdl.handle.net/20.500.11937/88713 10.1109/EAIS51927.2022.9787730 English ieee.org restricted
spellingShingle 4601 - Applied computing
4602 - Artificial intelligence
4603 - Computer vision and multimedia computation
4611 - Machine learning
0915 - Interdisciplinary Engineering
Vice, Jordan
Khan, Masood
Tan, Tele
Yanushkevich, Svetlana
Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title_full Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title_fullStr Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title_full_unstemmed Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title_short Dynamic Hybrid Learning for Improving Facial Expression Classifier Reliability
title_sort dynamic hybrid learning for improving facial expression classifier reliability
topic 4601 - Applied computing
4602 - Artificial intelligence
4603 - Computer vision and multimedia computation
4611 - Machine learning
0915 - Interdisciplinary Engineering
url http://hdl.handle.net/20.500.11937/88713