Deep Learning based Emotion Classification with Temporal Pupillometry Sequences

Rafique, Sidra; Kanwal, Nadia; Samar Ansari, Mohammad; Asghar, Mamoona; Akhtar, Zuhair

Abstract

In the recent era, automatic systems are the necessity of science. Systems for recognizing human emotions have gained popularity in various areas of knowledge specifically psychologists and psycho-physiologists. The interaction of the human-computer using physiological signals is the precise parameter for the recognition of emotion. However, pupillometry was used in this study as an unintentional direct brain response to capture human emotions using in-depth learning. Deep learning concepts using LSTM (Long Short Term Memory) were used in this study to classify emotions. Time series data for two emotions i.e. disgust and fear were used after the pre-treatment phase and subsequently proposed a classifier for the recognition of emotions.

Citation

Rafique, S., Kanwal, N., Samar Ansari, M., Asghar, M., & Akhtar, Z. (2021, December). Deep Learning based Emotion Classification with Temporal Pupillometry Sequences. Presented at 2021 International Conference on Electrical, Computer and Energy Technologies (ICECET), Cape Town, South Africa

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2021 International Conference on Electrical, Computer and Energy Technologies (ICECET)
• Start Date: Dec 9, 2021
• End Date: Dec 10, 2021
• Acceptance Date: Dec 9, 2021
• Publication Date: Dec 9, 2021
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Volume: 2021
• Series Title: 2021 International Conference on Electrical, Computer and Energy Technologies (ICECET)
• DOI: https://doi.org/10.1109/ICECET52533.2021.9698663
• Public URL: https://keele-repository.worktribe.com/output/423525
• Publisher URL: https://ieeexplore.ieee.org/document/9698663