How ARC Works

Existing solutions for a multiuser AR experience require dedicated devices or an internet connection. In ARKit's Multiuser Sessions, the AR World Map is synchronised among devices in a P2P fashion, but this implementation is exclusive to iOS users. On the other hand, ARCore's Cloud Anchors work both on Android and iOS devices; they require an active internet connection for data sharing though.

ARC's operating principle is simple: Let devices track their own position relative to a reference frame agreed upon beforehand. The reference frame position is established via image tracking on a flat surface. Finally, manage app connectivity over the local area network like any other LAN-based networked application.

• DETECTION: Devices are capable of tracking their surroundings with high precision. Therefore, each user must look around to detect the environment before joining a session. They must only agree on which Trackables correspond to particular points of interest in the physical environment.
• CONNECTION: Users connect to the same local network lobby. The current connection model is client-server, but session creation can be simplified using CaptainsMess' AutoConnect() functionality.
• CAPTURE: These points of interest can be flat objects - like a book, a drawing, or even a tablecloth - on a horizontal surface such as a table. They are then captured by the session host as top-down photos and will later act as markers in the physical environment. After confirmation, photos are shared with other connected users, as well as later joiners.
• SCANNING: Finally, all connected users must hover their device over the photographed objects. After detecting all markers, the app will display the AR content at the same position for everyone.

Once set up the app to perform the steps above, you can start developing your app content.

---

⟨ Prev	Next ⟩