A Real-Time and High Performance Posture Estimation System Based on Millimeter-Wave Radar

mmWave radar is a technique popular in industrial settings for applications such as production line tracking. It has recently been used for human detection and tracking within home health care as a potential application area. In this paper, we present a novel posture estimation system. We focus on categorisation of postures (standing, sitting, lying) for a single person in indoor environments. We establish a new point cloud dataset and show that, with simple algorithms employed, the system has over 99% accuracy for the field of view obtained when a single radar is placed above the participant.