Hambazi: Spatial Coordination Synthesis for Augmented Reality
summary
HAMBAZI implements well-organized replicated data types for multi-user AR using an escrow-style credit scheme plus a replication protocol and an SMT solver that derives minimal locks from linear-arithmetic constraints. It runs on off-the-shelf Android devices (tested up to 7), preserves integrity by blocking or queueing conflicting moves, tolerates crashes via aggregated recovery of credits and supervertex grouping, and reports latency reductions of 30.5–88.4% and location staleness drops of 35.6–75.6% versus three baselines.
AI-written summary · about
chapters (9)
- 0:00Spatial coordination in AR
- 1:23Multi-user AR coordination challenges
- 3:10Prior coordination limitations
- 4:55Spatial coordination synthesis
- 7:01Escrow mechanism implementation
- 9:33Credit-based failure handling
- 12:05SMT solvers for conflicts
- 13:08Experimental results
- 14:34Summary and future
related talks
38mDesigning Augmented Reality Systems for Speed
19mSynchronizAR: Instant Synchronization for Spontaneous and Spatial Collaborations in Augmented ...
01h 33mAugmenting Reality in Haskell
28mBuilding AR Experiences on Android
16mConnecting us to Each Other Means Connecting AR Apps to Each Other
10mSpatialstrates: Cross-Reality Collaboration through Spatial Hypermedia
01h 16mAugmented Reality For The Masses
18m4 Keys to Augmented Reality’s Future