Abnormal Event Detection Using Microsoft Kinect in a Smart Home

Abstract

In this paper, we present a continuous deep learning model for fall detection using Microsoft Kinect. The input include pre-processed high-resolution RGB images, depth images collected by a Kinect and optical flow images. We combine several deep learning structures including convolutional neural networks and long short-term memory networks for continuous human fallen detection. Finally, we present experimental results to demonstrate the performance and utility of our approach.