With the rapid advancement and widespread adoption of virtual reality (VR) technologies, the importance of accurate user identification within these platforms has gained importance for the security and privacy content. This thesis explores the potential of using human motion data as a biometric identifier within the games played in VR environments. Extensive studies involving 60 users were conducted to understand if head and hand movements can be distinguishing to identify the participants within multiple VR sessions. This research demonstrated that we can reliably identify participants with up to 90% accuracy using head and hand motion data as biometric markers. In this thesis, the movement data of 60 VR users were separated into two groups by playing one slow and one fast game with two different orders between four different VR games: Forklift Simulator, Beat Saber, Medal of Honor, and Cooking Simulator. The slow games that each participant played were the Forklift Simulator or Cooking Simulator and the fast games Beat Saber or Medal of Honor. While group one was playing Cooking Simulator and Beat Saber, group two played Forklift Simulator and Medal of Honor. The order has also changed; order one played the slow game first and order two played the fast game first. We achieved high identification accuracy with the movement data recordings thanks to this dual-game approach which allowed us to capture a wide range of movement patterns.
With the rapid advancement and widespread adoption of virtual reality (VR) technologies, the importance of accurate user identification within these platforms has gained importance for the security and privacy content. This thesis explores the potential of using human motion data as a biometric identifier within the games played in VR environments. Extensive studies involving 60 users were conducted to understand if head and hand movements can be distinguishing to identify the participants within multiple VR sessions. This research demonstrated that we can reliably identify participants with up to 90% accuracy using head and hand motion data as biometric markers. In this thesis, the movement data of 60 VR users were separated into two groups by playing one slow and one fast game with two different orders between four different VR games: Forklift Simulator, Beat Saber, Medal of Honor, and Cooking Simulator. The slow games that each participant played were the Forklift Simulator or Cooking Simulator and the fast games Beat Saber or Medal of Honor. While group one was playing Cooking Simulator and Beat Saber, group two played Forklift Simulator and Medal of Honor. The order has also changed; order one played the slow game first and order two played the fast game first. We achieved high identification accuracy with the movement data recordings thanks to this dual-game approach which allowed us to capture a wide range of movement patterns.
VR User Identification from Movement Analysis
KAYA, BEDRIYE GULCE
2023/2024
Abstract
With the rapid advancement and widespread adoption of virtual reality (VR) technologies, the importance of accurate user identification within these platforms has gained importance for the security and privacy content. This thesis explores the potential of using human motion data as a biometric identifier within the games played in VR environments. Extensive studies involving 60 users were conducted to understand if head and hand movements can be distinguishing to identify the participants within multiple VR sessions. This research demonstrated that we can reliably identify participants with up to 90% accuracy using head and hand motion data as biometric markers. In this thesis, the movement data of 60 VR users were separated into two groups by playing one slow and one fast game with two different orders between four different VR games: Forklift Simulator, Beat Saber, Medal of Honor, and Cooking Simulator. The slow games that each participant played were the Forklift Simulator or Cooking Simulator and the fast games Beat Saber or Medal of Honor. While group one was playing Cooking Simulator and Beat Saber, group two played Forklift Simulator and Medal of Honor. The order has also changed; order one played the slow game first and order two played the fast game first. We achieved high identification accuracy with the movement data recordings thanks to this dual-game approach which allowed us to capture a wide range of movement patterns.File | Dimensione | Formato | |
---|---|---|---|
Kaya_BedriyeGulce.pdf
accesso aperto
Dimensione
10.65 MB
Formato
Adobe PDF
|
10.65 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/65971