Unsupervised clustering of people from 'skeleton' data

Adrian Ball, David C. Rye, Fabio Ramos, Mari Velonaki

Abstract

This paper investigates the possibility of recognising individual persons from their walking gait using three-dimensional 'skeleton' data from an inexpensive consumer-level sensor, the Microsoft 'Kinect'. In an experimental pilot study it is shown that the K-means algorithm - as a candidate unsupervised clustering algorithm - is able to cluster gait samples from four persons with a nett accuracy of 43.6%.
BibTeX
@inproceedings{ball2012unsupervised,
  author = {Adrian Ball and David C. Rye and Fabio Ramos and Mari Velonaki},
  title = {{Unsupervised clustering of people from 'skeleton' data}},
  booktitle = {HRI},
  year = {2012},
  doi = {10.1145/2157689.2157767},
}

Loading related content...

Figure