Immersive Interaction Research Group
Our researches focus on embodied interactions, i.e. involving users through full-body movements to achieve new classes of tasks or activities not feasible with traditional human-computer interfaces. It can be employed for training and rehabilitation or for the evaluation of potentially complex environments. We also aim to better understand and mitigate cybersickness as it is still seriously hampering XR adoption. In collaboration with EPFL-LNCO we conduct experiments to assess the human embodiment sensitivity to various factors so that we can propose human-centered algorithms for full-body interaction. Independently we also investigated how to impersonate a full range of 3D characters (performance animation). Finally, our Know-how in human posture control has been used for proposing new 3D sketching tools for animators.
June 2024 : our most recent journal papers are visible in Open Access:
- IEEE TVCG “The Least Increasing Aversion (LIA) Protocol: Illustration on Identifying Individual Susceptibility to Cybersickness Triggers“. This new methodology is proposed to assess the impact of experimental factors causing potentially some discomfort, while minimizing the number of sessions.
- IEEEVR24/ TVCG “Who says you are so sick? An investigation on individual susceptibility to cybersickness triggers using EEG, EGG and ECG“. The objective data analysis of this quite rich exploratory study uncovers new knowledge about individual susceptibility to rotation axes.
- IEEEVR24/ TVCG “In case of doubt, one follows one’s self: the implicit guidance of the embodied self-avatar”. We show that the Self-Avatar Follower Effect also happens when the participant is unsure about what to do while performing large movements.
- Journal of Neural Engineering “Customizing the human-avatar mapping based on EEG error related potentials during avatar-based interaction“. This study demonstrates the feasibility of a closed-loop interaction based on the detection of an ErrP-based signal triggered by a Break in Embodiement.
- PLOS-One “EEG signature of breaks in embodiment in VR“. Our EEG data analysis show the occurrence of error-related potentials linked to observation of a Break in Embodiment event (BiE).
- PLOS-One “Avatar error in your favor: Embodied avatars can fix users’ mistakes without them noticing“. Through this controlled experiment, we showed an interesting asymmetry in embodied users behavior, i.e. that we notice much less when our avatar fix our mistake than when they introducing one.
