Vision-based Fall Detection: Methods and Challenges

Abstract

Fall accidents cause great harm to the elderly. Fall detection becomes an important research topic as the elderly population grows. In the research field of signal processing, many fall detection methods have been proposed, such as fall detection based on audio signal processing, radar, pressure sensor, acceleration sensor or computer vision. This paper briefly reviews vision-based fall detection, summarizes the advantages and disadvantages of recent methods, and discusses future topics.