Title :
Unsupervised clustering of people from ‘skeleton’ data
Author :
Ball, Adrian ; Rye, David ; Ramos, Fabio ; Velonaki, Mari
Author_Institution :
Centre for Social Robot., Univ. of Sydney, Sydney, NSW, Australia
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%.
Keywords :
gait analysis; human-robot interaction; interactive devices; object recognition; pattern clustering; Microsoft Kinect; candidate unsupervised clustering algorithm; consumer-level sensor; gait sample clustering; individual person recognition; k-means algorithm; people unsupervised clustering; three-dimensional skeleton data; walking gait; Clustering algorithms; Humans; Legged locomotion; Pattern recognition; Signal processing algorithms; Skeleton; Gait analysis; HRI; unsupervised clustering;
Conference_Titel :
Human-Robot Interaction (HRI), 2012 7th ACM/IEEE International Conference on
Conference_Location :
Boston, MA
Print_ISBN :
978-1-4503-1063-5
Electronic_ISBN :
2167-2121
