MOTION: ML-Assisted On-Device Low-Latency Motion Recognition

TL;DR

This paper introduces an efficient on-device motion recognition system using lightweight ML models and AutoML feature extraction on wearable sensors, enabling real-time gesture detection for medical monitoring.

Contribution

It demonstrates the feasibility of real-time, low-latency motion recognition on wearable devices using AutoML and lightweight neural networks, optimized for medical applications.

Findings

Neural networks achieved the best accuracy-latency trade-off.

Real-time gesture recognition was successfully implemented on WeBe Band.

The system is suitable for medical monitoring with fast response requirements.

Abstract

The use of tiny devices capable of low-latency gesture recognition is gaining momentum in everyday human-computer interaction and especially in medical monitoring fields. Embedded solutions such as fall detection, rehabilitation tracking, and patient supervision require fast and efficient tracking of movements while avoiding unwanted false alarms. This study presents an efficient solution on how to build very efficient motion-based models only using triaxial accelerometer sensors. We explore the capability of the AutoML pipelines to extract the most important features from the data segments. This approach also involves training multiple lightweight machine learning algorithms using the extracted features. We use WeBe Band, a multi-sensor wearable device that is equipped with a powerful enough MCU to effectively perform gesture recognition entirely on the device. Of the models explored, we found that the neural network provided the best balance between accuracy, latency, and memory use. Our results also demonstrate that reliable real-time gesture recognition can be achieved in WeBe Band, with great potential for real-time medical monitoring solutions that require a secure and fast response time.