Fall Detection With Multiple Cameras: An Occlusion-Resistant Method Based on 3-D Silhouette Vertical Distribution

TLDR

Elderly falls at home are rising as populations age, especially for those living alone, creating a growing need for emergency response. The authors propose a multi‑camera system that reconstructs a person’s 3‑D shape to detect falls at home. Fall detection is performed by analyzing the vertical distribution of volume and triggering an alarm when most of the distribution is abnormally close to the floor for a set duration. In validation with 24 realistic scenarios, the method achieved ≥99.7 % sensitivity and specificity with four or more cameras, and a GPU implementation ran in real time at 10 fps with eight cameras and 16 fps with three cameras.

Abstract

According to the demographic evolution in industrialized countries, more and more elderly people will experience falls at home and will require emergency services. The main problem comes from fall-prone elderly living alone at home. To resolve this lack of safety, we propose a new method to detect falls at home, based on a multiple-cameras network for reconstructing the 3-D shape of people. Fall events are detected by analyzing the volume distribution along the vertical axis, and an alarm is triggered when the major part of this distribution is abnormally near the floor during a predefined period of time, which implies that a person has fallen on the floor. This method was validated with videos of a healthy subject who performed 24 realistic scenarios showing 22 fall events and 24 cofounding events (11 crouching position, 9 sitting position, and 4 lying on a sofa position) under several camera configurations, and achieved 99.7% sensitivity and specificity or better with four cameras or more. A real-time implementation using a graphic processing unit (GPU) reached 10 frames per second (fps) with 8 cameras, and 16 fps with 3 cameras.