Exercise Motion Classification from Large-Scale Wearable Sensor Data Using Convolutional Neural Networks

TL;DR

This paper presents a CNN-based approach to classify 50 gym exercises from large-scale wearable sensor data, achieving over 92% accuracy by transforming time-series data into images for automatic feature extraction.

Contribution

It introduces a novel method of formatting sensor time-series data as images for CNN classification and compares different CNN architectures for exercise recognition.

Findings

Achieved 92.1% accuracy in classifying 50 exercises

Found optimal image formatting improves CNN performance

Compared various CNN architectures for best results

Abstract

The ability to accurately identify human activities is essential for developing automatic rehabilitation and sports training systems. In this paper, large-scale exercise motion data obtained from a forearm-worn wearable sensor are classified with a convolutional neural network (CNN). Time-series data consisting of accelerometer and orientation measurements are formatted as images, allowing the CNN to automatically extract discriminative features. A comparative study on the effects of image formatting and different CNN architectures is also presented. The best performing configuration classifies 50 gym exercises with 92.1% accuracy.