A Demonstration of Inferring Identity using Accelerometers in Television Remote Controls

This demonstration shows how accelerometers and buttonpress loggers embedded in a television remote control can be used to distinguish different participants based on the unique way each person wields the remote. This is a live demonstration of the inference technology presented in the paper Inferring Identity using Accelerometers in Television Remote Controls, which appears as a full paper in Pervasive 2009. Unlike the full paper, which presents results from five 3-week data sets collected from real homes, this demonstration simplifies the machine learning and instead does user-classification with real-time training data collected from visitors to the demo kiosk.