Designing and Analyzing Virtual Avatar Based on Rigid-Body Tracking in Immersive Virtual Environments

This study proposes a novel method for virtual avatar applications based on rigid-body tracking in immersive virtual environments (IVEs). This method aims to design a method to efficiently and accurately estimate motions defined with few conditions. For this purpose, we design a virtual avatar motion process that proceeds from rigid-body tracking to motion estimation using Final IK (inverse kinematics). This study uses optical motion capture equipment based on markers for rigid-body tracking and calculating the movement of a defined virtual avatar based on the transformation information of the precisely tracked rigid-body object. To analyze the effectiveness of the proposed method, a comparative analysis of equipment and environment with existing research on animation fidelity in self-avatars is performed. In addition, two upper body-centered actions (simultaneous two-hand motion and separate two-hand motion) are presented to analyze latent factors of virtual avatar embodiment. A comparative survey experiment using the virtual embodiment questionnaire (VEQ) is conducted based on full-body tracking and the proposed rigid-body tracking-based method. Furthermore, to compare and analyze users’ realism in an environment where they coexist with virtual avatars in IVE, a survey experiment using a presence questionnaire is conducted. Therefore, we confirmed that similar avatar motion generation is possible compared to full-body tracking even with fewer tracking targets (e.g. type of tracking device or number of markers), inputs, and processing steps. Additionally, it is possible to design a virtual avatar that provides satisfactory realism in IVE.