| dc.description.abstract | In this dissertation, a novel collaborative augmented reality (AR) based framework was developed that will allow two users who are geographically separated to have naturalistic interactions with each other and their environment. The goal is to allow a local user to be present in their remote counterpart’s environment via their photorealistic avatar and interact with in a way that would be akin to physically being present from the perspective of the remote user and vice versa. This includes not only being able to map the user’s object interactions to their avatar, but also preserve other subtle nonverbal behaviors, such as one’s hand gestures, overall body pose, proxemics behavior and locomotion patterns, aspects that are not captured by current AR frameworks. The user’s actions, however, cannot be directly mapped onto their avatar without causing an interaction error, primarily due to the difference in spatial configuration between the user’s and their avatar’s environment. In this work, a mapping framework is developed that includes algorithms and techniques to rectify these interaction errors while at the same time preserve the nonverbal aspects of the interaction. Specifically, we categorized the mapping problem into two components, i.e., the redirection of the kinematic structure of the user’s body and redirection of the locomotion of the user, and solved specific problems related to these components in such a way that they can be integrated into an overall collaborative AR framework for realistic, nonverbal behavior preserving interaction. This ability to simulate naturalistic 3D visits from remote locations have many applications in consumer, workplace, and therapeutic domains. In this dissertation, as an application of the proposed framework, we will simulate 3D visits for an important and growing population group: older adults. The goal is to find aspects of the framework that they would like to see improved and activities that they would prefer to help them to connect with their loved ones. Additional issues such as the system design, implementation, and the logistics of conducting such an activity are also considered. The information from this pilot study will be helpful for conducting larger studies using this framework to mitigate loneliness in older adults. | |
