The future of the Web lies in integration of Web content with the real-world through technologies such as augmented-reality. This overlay of virtual content on top of the physical world, called the Spatial Web, holds promise for dramatically changing many applications. For example, imagine being able to access an individual’s home page by looking at them, seeing directions from a mapping application shown overlaid on the actual roads, or seeing control interfaces for your appliances simply by looking at them. The Spatial Web will impact applications from navigation, to engineering, education and more.
While the vision is promising, the reality is that Spatial Web applications are currently difficult to create. Early applications such as Google Maps AR interface are monolithic designs with a single goal. A broad range of new tools and infrastructure are needed to make the Spatial Web "real" and to make it easy for developers to create and publish Spatial Web content and applications. The goal of this project is to develop a coherent architecture for the Spatial Web. This includes tools and algorithms for each step in the process of generating a Spatial Web view for the user.
The proposed research directly addresses the key challenges associated with creating Spatial Web applications today. For example, the PIs plan to develop novel localization and scene understanding tools to help accurately position the user and the objects in their view. The planned research will look at reducing deployment costs and accommodating greater heterogeneity in localization frameworks. The PIs also plan to design a system for spatial content management to simplify the process of finding relevant spatial content and propagating content updates to clients. Key challenges the PI address in this lookup system include scalably handling diverse spatial queries, enabling administrative control and management of Spatial Web lookups, and providing a low-latency publish-subscribe interface to enable real-time discovery and update of in-view Spatial Web content. Finally, the PIs plan to develop content encoding and transmission protocols that are particularly well-suited to the low-latency requirements of the Spatial Web. This includes new video codec and video transmission schemes that are optimized for low-latency interaction, designs for congestion control protocols that minimize queuing delay in environments with cross-traffic, and new QoE metrics to evaluate the performance of Spatial Web content delivery.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
