Two-stage Fall Events Classification with Human Skeleton Data

TL;DR

This paper introduces a two-stage deep learning approach using human skeleton data for multi-class fall event classification, enhancing privacy and accuracy over existing binary methods in healthcare applications.

Contribution

It presents a novel two-stage classification framework that first filters no-fall events and then classifies multiple fall types using privacy-preserving skeleton features.

Findings

Outperforms state-of-the-art methods on UP-Fall dataset

Effectively classifies five types of fall events

Reduces privacy concerns with skeleton data

Abstract

Fall detection and classification become an imper- ative problem for healthcare applications particularity with the increasingly ageing population. Currently, most of the fall clas- sification algorithms provide binary fall or no-fall classification. For better healthcare, it is thus not enough to do binary fall classification but to extend it to multiple fall events classification. In this work, we utilize the privacy mitigating human skeleton data for multiple fall events classification. The skeleton features are extracted from the original RGB images to not only mitigate the personal privacy, but also to reduce the impact of the dynamic illuminations. The proposed fall events classification method is divided into two stages. In the first stage, the model is trained to achieve the binary classification to filter out the no-fall events. Then, in the second stage, the deep neural network (DNN) model is trained to further classify the five types of fall events. In order to confirm the efficiency of the proposed method, the experiments on the UP-Fall dataset outperform the state-of-the-art.