Fall detection using multimodal data

TL;DR

This paper presents a multimodal fall detection system using sensor and camera data, achieving superior accuracy over existing methods on the UP-Fall Detection Dataset.

Contribution

It introduces novel feature extraction techniques and models for fall detection using multimodal data, outperforming state-of-the-art approaches.

Findings

Our methods surpass existing approaches in accuracy, precision, recall, and F1 score.

The use of multimodal data improves fall detection performance.

The proposed techniques demonstrate robustness across diverse sensor and camera inputs.

Abstract

In recent years, the occurrence of falls has increased and has had detrimental effects on older adults. Therefore, various machine learning approaches and datasets have been introduced to construct an efficient fall detection algorithm for the social community. This paper studies the fall detection problem based on a large public dataset, namely the UP-Fall Detection Dataset. This dataset was collected from a dozen of volunteers using different sensors and two cameras. We propose several techniques to obtain valuable features from these sensors and cameras and then construct suitable models for the main problem. The experimental results show that our proposed methods can bypass the state-of-the-art methods on this dataset in terms of accuracy, precision, recall, and F1 score.