Depth Sensor Based Skeletal Tracking Evaluation for Fall Detection Systems

Falls are very common in elderly due to various physical constraints. Since falls may cause serious injury and even death, fall detection systems are very important, especially when the victim is alone at home or is unable to seek regular/timely medical assistance. In this paper, development of a fall detection system based on Kinect sensor is evaluated. Microsoft Kinect is a low cost RGB-D sensor and it has the ability to track joint positions which could prove useful as a sophisticated tool for fall detection. In this study the potential of Kinect for application in fall detection has been investigated.