Master's Thesis: Detecting Immersion with Machine Learning

Immersion Detection in 2D media by means of Physiological data, a Master’s dissertation submitted in order to obtain the academic degree of Master of Science in Computer Science Engineering.

During my master’s dissertation, I explored the potential of Machine Learning (ML) in identifying connections between immersion in multimedia content and physiological signals, primarily focusing on modeling with electroencephalograms (EEGs).

Main Contributions:

• Objective condition classification: The study showed that EEG data can effectively distinguish between cases where users watch content in objectively different settings.
• Subjective rating classification: The research also revealed that EEG data can accurately predict when a user will rate an experience as immersive.

Implications

My master thesis emphasizes the feasibility of modeling cognitive processes such as immersion using EEGs. These findings could enable personalized immersive experiences in emerging multimedia formats and have implications for various industries, including entertainment, healthcare, and education.

Acknowledgements

I have to thank the various researchers at Ghent University who helped me during my master’s dissertation: Prof. dr. ir. Maria Torres Vega, Ir. Sam Van Damme, Ir. Mohammad Javad Sameri and Prof. dr. ir. Filip De Turck.
I’d also like to express my gratitude to Dr. Anne-Flore Perrin, for the data and prior research in the field of quality assessment of immersive media.