Demonstrating CrowdInside: A system for the automatic construction of indoor floor-plans

The existence of a worldwide indoor floor-plans database can lead to significant growth in pervasive computing, especially for indoor environments. In this demonstration, we show CrowdInside: a crowdsourcing-based system for the automatic construction of buildings floor-plans. CrowdInside leverages the smart phones sensors that are ubiquitously available with humans who use a building to automatically and transparently construct accurate motion traces. CrowdInside processes the collected motion traces from different visitors of a building to detect its overall floor-plan shape as well as higher level semantics such as number of rooms, their shapes and locations, and corridors shapes along with a variety of points of interest in the environment. The goal of this demo is to show how the accurate motion traces are constructed as well as how the building floor-plan can be automatically generated.