University of Twente Student Theses


Posture as stress indicator in eSports

Wolbers, P.W. (2022) Posture as stress indicator in eSports.

[img] PDF
Abstract:eSports has grown significantly in the past few years. Whilst casual gaming has a global market cap of 200 billion $, professional tournaments can have prize pools of over 34 million $. Because of the very high stakes and drive to win, stress is apparent in both casual and professional players. However, research in eSports is very young, meaning not much is known of the effects of stress and gaming. The premise of this project was to predict stress, by comparing individual stress response to their physical posture. The idea is to use posture to predict stress, as this would keep measurements non-intrusive, such that individual performance can be the best. From background research, multiple options for stress measurement and pose estimation were found. As priority was given to non-intrusiveness, only a few options remained. On the stress side, the Empatica E4 was chosen for its ease of use and SDK integration. For posture, OpenPose was chosen for its versatility. The resulting prototype used EDA peaks as measure of stress, specifically, the number of peaks in 30s. Posture was classified with LSTM machine learning, as other options had an inherent bias. During the evaluation, the prototype was tested with 3 college level professional players. Due to unforeseen circumstances and bad preparation, data from these tests were unusable. It was therefore decided to do the test on myself, for a total of three matches. Results showed no significant correlation between stress and posture. Certain moments did show correlation between the two, however, this was only visible during very stressful moments. Furthermore, these results were derived from a casual player, implying that a professional player will most likely show a lesser physical response to stressful situations. Future work should look at different classification of stress and a different configuration of OpenPose.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Creative Technology BSc (50447)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page