Recognising Dutch Sign Language Signs Using the Kinect: Hidden Markov Models vs Dynamic Time Warping

Keywords
No Thumbnail Available
Issue Date
2017-06-29
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This paper compares the abilities of Hidden Markov Models and Dynamic Time Warping to classify Dutch Sign Language gestures with few examples. As communication with computers becomes more important researchers need to know the methods they should use to maximise their results given their data. 9 subjects recorded gestures for this experiment. With enough examples Hidden Markov Models can give you a good classification, but with few examples Dynamic Time Warping proves to be a better classifier.
Description
Citation
Faculty
Faculteit der Sociale Wetenschappen