Partially observable markov decision process (POMDP) technologies for sign language based human-computer interaction

Ong, Sylvie C. W., Hsu, David, Lee, Wee Sun and Kurniawati, Hanna (2009). Partially observable markov decision process (POMDP) technologies for sign language based human-computer interaction. In: Constantine Stephanidis, Universal Access in Human-Computer Interaction: Applications and Services. 5th International Conference on Universal Access in Human-Computer Interaction (UAHCI 2009), San Diego, United States, (577-586). 19-24 July 2009. doi:10.1007/978-3-642-02713-0_61


Author Ong, Sylvie C. W.
Hsu, David
Lee, Wee Sun
Kurniawati, Hanna
Title of paper Partially observable markov decision process (POMDP) technologies for sign language based human-computer interaction
Conference name 5th International Conference on Universal Access in Human-Computer Interaction (UAHCI 2009)
Conference location San Diego, United States
Conference dates 19-24 July 2009
Proceedings title Universal Access in Human-Computer Interaction: Applications and Services   Check publisher's open access policy
Journal name Lecture Notes in Computer Science   Check publisher's open access policy
Place of Publication Heidelberg, Germany
Publisher Springer
Publication Year 2009
Sub-type Fully published paper
DOI 10.1007/978-3-642-02713-0_61
Open Access Status
ISBN 9783642027123
9783642027130
ISSN 0302-9743
1611-3349
Editor Constantine Stephanidis
Volume 5616
Issue Part III
Start page 577
End page 586
Total pages 10
Language eng
Abstract/Summary Sign language (SL) recognition modules in human-computer interaction systems need to be both fast and reliable. In cases where multiple sets of features are extracted from the SL data, the recognition system can speed up processing by taking only a subset of extracted features as its input. However, this should not be realised at the expense of a drop in recognition accuracy. By training different recognizers for different subsets of features, we can formulate the problem as the task of planning the sequence of recognizer actions to apply to SL data, while accounting for the trade-off between recognition speed and accuracy. Partially observable Markov decision processes (POMDPs) provide a principled mathematical framework for such planning problems. A POMDP explicitly models the probabilities of observing various outputs from the individual recognizers and thus maintains a probability distribution (or belief) over the set of possible SL input sentences. It then computes a policy that maps every belief to an action. This allows the system to select actions in real-time during online policy execution, adapting its behaviour according to the observations encountered. We illustrate the POMDP approach with a simple sentence recognition problem and show in experiments the advantages of this approach over “fixed action” systems that do not adapt their behaviour in real-time.
Keyword Robotics
Sign language recognition
Human-computer interaction
Planning under uncertainty
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status Non-UQ
Additional Notes The 5th International Conference on Universal Access in Human-Computer Interaction was part of the The 13th International Conference on Human–Computer Interaction (HCI International 2009) that was held in San Diego, California, USA, July 19–24, 2009. HCI International 2009 was held jointly with the Symposium on Human Interface (Japan) 2009, the 8th International Conference on Engineering Psychology and Cognitive Ergonomics, the 5th International Conference on Universal Access in Human-Computer Interaction, the Third International Conference on Virtual and Mixed Reality, the Third International Conference on Internationalization, Design and Global Development, the Third International Conference on Online Communities and Social Computing, the 5th International Conference on Augmented Cognition, the Second International Conference on Digital Human Modeling, and the First International Conference on Human Centered Design.

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Wed, 16 Apr 2014, 20:53:04 EST by Ms Kimberley Nunes on behalf of Centre for Medical Diagnostic Technologies in Qld