Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a 15-minute conference presentation introducing Hambazi, a novel approach to spatial coordination synthesis for multi-user augmented reality applications. Learn how researchers from UC Riverside, University of Michigan, and UC Santa Cruz address the critical challenge of maintaining consistent, low-latency virtual object coordination across multiple AR users while preserving spatial integrity properties. Discover the concept of "well-organized" replicated data types that guarantee both quick convergence and spatial constraints, such as preventing virtual objects from violating physical boundaries or privacy zones. Examine the innovative credit scheme and replication protocol that enables local execution while maintaining correctness, and understand how constraint solving automatically derives conflicting actions to synthesize custom coordination for specific AR environments. Review the evaluation results demonstrating 30.5-88.4% latency reduction and 35.6-75.6% location staleness reduction compared to baseline approaches across various device configurations, AR environments, request loads, and network conditions, with implementation tested on off-the-shelf Android AR devices.
