Citation

Gan, Yip Fong and Goh, Pey Yun and Chong, Lee Ying and Tan, Shing Chiang (2024) A Comparative Study: Facial Emotion Recognition by Using Deep Learning. In: 2024 12th International Conference on Information and Communication Technology (ICoICT), 07-08 August 2024, Bandung, Indonesia.

Text A Comparative Study_ Facial Emotion Recognition by Using Deep Learning.pdf - Published Version Restricted to Repository staff only Download (667kB)

Abstract

Facial emotion recognition (FER) has garnered significant attention due to its applications in human-computer interaction, healthcare, psychology, and entertainment. Deep learning algorithms have been widely applied and improve the performance of FER. However, some critical gaps in FER, including the lack of standardized comparative platforms, limited used of evaluation metrics may have hinder the comprehensive understanding of various deep learning algorithms in FER. This study proposed some useful guidelines which include a standardized platform, more evaluation metrics and explored the impact of transfer learning and input size on FER. These datasets: CK, FER aligned, FER aligned + CK, RAF-DB, and AffectNet are used in the experiment. The research used the cloud-based platform, i.e. Kaggle as the standardized platform and the code was released through GitHub. Through the suggested guidelines for evaluation, some interesting insights such as the performance of each different emotion for each dataset and each deep learning algorithm, the impact of different input sizes and transfer learning are revealed.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Deep learning algorithms
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	03 Dec 2024 00:23
Last Modified:	03 Dec 2024 00:23
URII:	http://shdl.mmu.edu.my/id/eprint/13140

Downloads

Edit (login required)