Gait Classification on Smartphones

Start Year:

2011

End Year:

2013

Smartphones have a wealth of sensors in them. To improve ubiquitous application, we use smartphone accelerometers to measure information about gait (how a person walks). We've developed the RRACE algorithm that can sense a user's cadence (step frequency) without any training on the person. We also used machine learning to classify different gaits to provide input to an exercise game or other active application. We have working implementations of both systems that run on Android.

SPIN People:

Oliver Schneider

Mike Wu

Idin Karuei

Publications:

Schneider, O., "Using Gait as an Input Modality for Mobile Exercise Games,", M.Sc. Thesis, University of British Columbia, 2012.

Schneider, O. et al “Real-time Gait Classification for Persuasive Smartphone Apps: Structuring the Literature and Pushing the Limits,” in Proc. of Int'l Conf in Intelligent User Interfaces (IUI '13), Santa Monica, USA, pp. 161-172 2013.

Tags:

machine learning

mobile

movement sensing