Paper Info
Title
Real-Time Motion Tracking for Mobile Augmented/Virtual Reality Using Adaptive Visual-Inertial Fusion.
Abstract
In mobile augmented/virtual reality (AR/VR), real-time 6-Degree of Freedom (DoF) motion tracking is essential for the registration between virtual scenes and the real world. However, due to the limited computational capacity of mobile terminals today, the latency between consecutive arriving poses would damage the user experience in mobile AR/VR. Thus, a visual-inertial based real-time motion tracking for mobile AR/VR is proposed in this paper. By means of high frequency and passive outputs from the inertial sensor, the real-time performance of arriving poses for mobile AR/VR is achieved. In addition, to alleviate the jitter phenomenon during the visual-inertial fusion, an adaptive filter framework is established to cope with different motion situations automatically, enabling the real-time 6-DoF motion tracking by balancing the jitter and latency. Besides, the robustness of the traditional visual-only based motion tracking is enhanced, giving rise to a better mobile AR/VR performance when motion blur is encountered. Finally, experiments are carried out to demonstrate the proposed method, and the results show that this work is capable of providing a smooth and robust 6-DoF motion tracking for mobile AR/VR in real-time.
Year
DOI
Venue
2017
10.3390/s17051037
SENSORS
Keywords
Field
DocType
real-time motion tracking,adaptive filter,visual-inertial fusion,mobile AR/VR,pose estimation
Computer vision,User experience design,Virtual reality,Motion blur,Pose,Robustness (computer science),Artificial intelligence,Adaptive filter,Jitter,Engineering,Match moving
Journal
Volume
Issue
Citations 
17
5
5
PageRank 
References 
Authors
0.46
25
4
Name
Order
Citations
PageRank
Wei Fang133919.89
Lianyu Zheng292.88
Huanjun Deng350.46
Hongbo Zhang432233.83