X. Qiao, P. Ren, G. Nan, L. Liu, S. Dustdar, J. Chen:
"Mobile Web Augmented Reality in 5G and Beyond: Challenges, Opportunities, and Future Directions";
The popularity of wearable devices and smartphones has fueled the development of Mobile Augmented Reality (MAR), which provides immersive experiences over the real world using techniques, such as computer vision and deep learning. However, the hardware-specific MAR is costly and heavy, and the App-based MAR requires an additional download and installation and it also lacks cross-platform ability. These limitations hamper the pervasive promotion of MAR. This paper argues that mobile Web AR (MWAR) holds the potential to become a practical and pervasive solution that can effectively scale to millions of end-users because MWAR can be developed as a lightweight, cross-platform, and low-cost solution for end-to-end delivery of MAR. The main challenges for making MWAR a reality lie in the low efficiency for dense computing in Web browsers, a large delay for real-time interactions over mobile networks, and the lack of standardization. The good news is that the newly emerging 5G and Beyond 5G (B5G) cellular networks can mitigate these issues to some extent via techniques such as network slicing, device-to-device communication, and mobile edge computing. In this paper, we first give an overview of the challenges and opportunities of MWAR in the 5G era. Then we describe our design and development of a generic service-oriented framework (called MWAR5) to provide a scalable, flexible, and easy to deploy MWAR solution. We evaluate the performance of our MWAR5 system in an actually deployed 5G trial network under the collaborative configurations, which shows encouraging results. Moreover, we also share the experiences and insights from our development and deployment, including some exciting future directions of MWAR over 5G and B5G networks.
5G; edge computing; augmented reality; mobile augmented reality; Web augmented reality
"Offizielle" elektronische Version der Publikation (entsprechend ihrem Digital Object Identifier - DOI)
Erstellt aus der Publikationsdatenbank der Technischen Universitšt Wien.