English

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

Computer Vision and Pattern Recognition 2017-07-25 v3 Machine Learning

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.

Keywords

Cite

@article{arxiv.1610.07031,
  title  = {Exercise Motion Classification from Large-Scale Wearable Sensor Data Using Convolutional Neural Networks},
  author = {Terry Taewoong Um and Vahid Babakeshizadeh and Dana Kulić},
  journal= {arXiv preprint arXiv:1610.07031},
  year   = {2017}
}

Comments

will appear in IROS2017

R2 v1 2026-06-22T16:28:25.940Z