Preliminary results on algorithms for multi-kinect trajectory fusion in a living lab

Abstract Everyday activity of an individual is related to his health status. In order to improve daily health monitoring at home, an indoor position tracking system has been designed. The latter is based on a network of depth cameras to detect and track people's position. The trajectories obtained from each camera are merged to reconstruct each individual's own entire trajectory within the apartment, from which home activities can be derived. In this way, the early detection of a change in daily activities of the elderly will highlight disabilities and loss of autonomy. Standard modules and software were implemented in the system architecture to integrate sensors and systems seamlessly to provide high flexibility and integration capacity for future developments. This system is meant to improve homecare health management for a better end of life at an affordable price for the community.