Activity recognition of assembly tasks using body-worn microphones and accelerometers

Jamie A Ward1, Paul Lukowicz, Gerhard Tröster

  • 1Swiss Federal Institute of Technology (ETH), Institute for Electronics, Zürich, Switzerland. ward@ife.ee.ethz.ch

Summary

This study introduces a method for wearable computers to recognize manual tasks using sound and motion sensors. Continuous activity recognition achieved 78% recall and 74% precision in a workshop setting.

Related Concept Videos