Crowdstream - Displaying Video across the Phones of Large Crowds at Concerts

Abstract

This report proposes a new, novel method for indoor positioning that is better suited than existing methods at providing relative positions in indoor venues. The research set out to investigate if it is possible to create a system that can be used to orchestrate a position-aware light show across the devices of an audience in a large indoor arena, using only an inexpensive, off-the-shelf camera and the devices themselves to determine their positions.

The resulting approach is cheaper, easier to set up, provides better relative accuracy and is less susceptible to interference than any other existing solution when used under specific conditions and assumptions. It is accompanied by a fully functional prototype of a system that can serve as a platform for further research and improvement. The prototype includes a server with a simple but extendable implementation of a new approach to Indoor Positioning Systems based on mobile devices and a camera, an Android client, and a administrator interface for managing the server.

Preliminary results show that the system is capable of detection, positioning and streaming with at least 500 devices in an indoor venue, and provides ideas and suggestions for further research and improvements.