Citation

Md Suhaidin, Siti Nabila and Mogan, Jashila Nair and Noviaristanti, Siska (2026) Gait-Based Emotion Recognition Using Enhanced Xception. In: 14th International Conference on Information and Communication Technology, ICoICT 2026, 30 July 2026 - 31 July 2026, Bandung.

Text 17.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The paper utilized deep learning techniques for the classification of human emotions based on the patterns of walking. The model used in this research has a modified version of the Xception model that can extract very detailed spatial and temporal features from the gait as well as from the four main emotions: i.e., angry, happy, neutral, and sad. The researchers utilized the E-Walk (EmotionWalk) dataset comprising raw videos that were later processed to obtain and label the key frames. The Xception model is then trained and tested using these frames. Amongst the many models that were tested, the Xception yielded a top classification accuracy of 99.47%, thus surpassing other contemporary models whose accuracies ranged from 56.00% to 95.00%. This victory shows how much better Xception is at recognizing emotion-associated walking patterns than any other. The study underlines the possibility of using walking as a non-invasive technology for the detection of mood states. Moreover, the comparison study provides information about the model's strength and the accuracy of the classifications, and it can be considered as a starting point for the future development of multimodal approaches. The work presented in this paper is a step forward in affective computing as it opens a new way of recognizing emotions through gait in a convenient and non-intrusive way, which can be applied in mental health assessment, surveillance, and human-computer interaction, for instance.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Emotion recognition, gait-based, Xception
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	02 Oct 2026 07:52
Last Modified:	02 Oct 2026 07:52
URII:	http://shdl.mmu.edu.my/id/eprint/16842

Downloads

Edit (login required)