Behavioral Authentication with Head-Tilt Based Locomotion for Metaverse
Zijun Mai, Yidang He, Jiana Feng, Huawei Tu, Jian Weng, BoYu Gao · 2023
In order to address issues with user authentication such as shoulder surfing attacks and authentication timeouts, we proposed an implicit authentication method for locomotion in virtual reality (VR). The method used head-tilt based locomotion for user authentication. We conducted a study to collect a dataset of biological behavior for head-tilt based locomotion. Random Forest and Decision Tree algorithms were used to train the model, comparing the fitting effects of the two algorithms at different distances (3 meters, 6 meters, 9 meters, and 12 meters). We examined the model's accuracy and discovered that it had a high accuracy in identifying one of ten users (above 95 %).