Since 2010 when Microsoft released the Kinect
sensor, the price of depth sensors has fallen,
enabling the adoption of previously high-end
technology for new application domains. The
first version of the VTT people tracking system
was built on a one sensor approach using these
cost-effective depth sensors. The networked
solution [1] in which a centralized server collects
and fuses the detection events signalled by the
sensor nodes has been demonstrated on many
occasions in the last couple of years and has
attracted interest. The novelty of our solution is
in the developed data analysis functionalities,
enabling behaviour modelling and leading to
intention awareness as well as to prediction
of people behaviour [3], [4]. The basic people
tracking system has been integrated with external
services: lighting control [2], interactive spaces,
customer monitoring and facility monitoring
services.