PolyVerse: An Edge Computing-Empowered Metaverse with Physical-to-Virtual Projection

Incorporating physical-world objects and information into the virtual world is a key aspect of enhancing immersiveness in the metaverse. However, existing solutions are neither efficient nor scalable when projecting enormous objects due to their reliance on expensive specialized equipment. To fill this gap, we propose PolyVerse, an edge computing-powered metaverse platform that supports practical Physical-to-Virtual (P2V) projection while preserving low costs. PolyVerse tackles two challenges in P2V. First, P2V inherently requires real-time projection of enormous physical objects, which is difficult to achieve without specialized equipment. To overcome this issue, we alternatively utilize edge computing-based methods to collect, process, and analyze large-scale physical-world data with cost-effective edge devices and AI models. Workloads are adaptively scheduled among edge devices through a heuristic algorithm to reduce latency. Second, metaverse users may encounter inconsistent states among multiple metaverse service providers, which can be caused by potential networking issues or malicious state tampering when projecting vast objects. To this end, we develop a blockchain-based metaverse management scheme among service providers to ensure a consistent view for users. States are initially constructed as a Metaverse State Tree that supports efficient accumulation, retrieval, and membership proof generation. The tree digests are further secured by blockchain consensus to ensure consistency. Finally, we develop a metaverse campus prototype where real-world pedestrians are projected into the virtual world in real-time. Evaluation shows that PolyVerse can project objects within an average latency of 250ms.