No abstract

Figure 1: The virtual interview experiment in different avatar control conditions with first-person view (FPV), third-person view (TPV), and real-world view (RWV): (a) Controller-based interview session, task 1: Answer the questions, (b) Controllerbased interview session, task 2: Route planning task, (c) Depth-sensor-based interview session, task 1: Answer the questions, (d) Depth-sensor-based interview session, task 2: Route planning task

Avatar-mediated collaboration in virtual environments is becoming more and more prevalent. However, current consumer systems are not suited to fully replicate real-world nonverbal communication. We present a novel avatar system for collaboration in virtual reality, which supports high levels of nonverbal expression by tracking behavior such as body movement, hand gesture, and facial expression. The system was built using camera tracking technology only. Therefore, in contrast to many other high-level tracking systems, it does not require users to wear additional trackers on their bodies. We compared our highly expressive system with a consumer setup extended with two body-worn trackers in a dyadic study. We investigated users’ performance, such as completion time and accuracy, as well as the presence and interpersonal attraction in a virtual charades game using an asymmetric control scheme. The results show that participants interacting with highly expressive avatars felt more social presence and attraction and exhibited better task performance than those interacting with partners represented using low-expressive avatars. Hence, we conclude that virtual reality avatar systems benefit from a higher level of nonverbal expressiveness, which can be achieved without additional body-worn trackers.

No abstract

No abstract

No abstract

Skeleton tracking based on multiple Kinects data fusion has been proved to have better accuracy and robustness than single Kinect. However, previous works did not consider the inconsistency of tracking accuracy in the tracking field of Kinect and the self-occlusion of human body in assembly operation, which are of vital importance to the fusion performance of the multiple Kinects data in assembly task simulation. In this work, we developed a multi-Kinect fusion algorithm to achieve robust full-body tracking in virtual reality (VR)-aided assembly simulation. Two reliability functions are first applied to evaluate the tracking confidences reflecting the impacts of the position-related error and the self-occlusion error on the tracked skeletons. Then, the tracking skeletons from multiple Kinects are fused based on weighted arithmetic average and generalized covariance intersection. To evaluate the tracking confidence, the ellipsoidal surface fitting was used to model the tracking accuracy distribution of Kinect, and the relations between the user-Kinect crossing angles and the influences of the self-occlusion on the tracking of different parts of body were studied. On the basis, the two reliability functions were developed. We implemented a prototype system leveraging six Kinects and applied the distributed computing in the system to improve the computing efficiency. Experiment results showed that the proposed algorithm has superior fusion performance compared to the peer works.

No abstract