Recognition of Hand Gestures Tracked by a Dataglove: Exploiting Hidden Markov Models Discriminative Training and Environment Description to Improve Recognition Performance

Vittorio Lippi, Emanuele Ruffaldi, Carlo Alberto Avizzano, Massimo Bergamasco

2009

Abstract

Sequence classification based on Hidden Markov Models (HMMs) is widely employed in gesture recognition. Usually, HMMs are trained to recognize sequences by adapting their parameters through the Baum-Welch algorithm, based on Maximum Likelihood (ML). Several articles have pointed out that ML can lead to poor discriminative performances among gestures. This happens because ML is not optimal for this purpose until the modellized process is actually an HMM. In this paper we present a gesture recognition system featuring a discriminative training algorithm based on Maximal Mutual Information (MMI) and the integration of environment information. The environment is described through a set of fuzzy clauses, on the basis of which a priori probabilities are computed. Adaptive systems such as unsupervised neural networks are used to build a codebook of symbols representing the hand’s states. An experiment on a set of meaningful gestures performed during the interaction with a virtual environment is then used to evaluate the performance of this solution.

References

  1. Melinda M. Cerney, J.M.V.: Gesture recognition in virtual environments: A review and framework for future development. Technical report, Iowa State University (2005)
  2. Brough, J. E., Schwartz, M., Gupta, S.K., Anand, D.K., Kavetsky, R., Pettersen, R.: Towards the development of a virtual environment-based training system for mechanical assembly operations. Virtual Real. 11 (2007) 189-206
  3. Moeslund, T.B.: Principal component analysis - an introduction. Technical Report Technical Report CVMT 01-02, ISSN 0906-6233, Aalborg University (2001)
  4. Mathworks: Matlab manual. Mathworks. (2004)
  5. Bishop, C.M.: Pattern Recognition and Machine Learning. Springer (2006)
  6. L.R. Rabiner, B.J.: An introduction to hidden markov models. IEEE ASSP Magazine 3 (1986) 4- 16
  7. Chow, Y.L.: Maximum mutual information estimation of hmm parameters for continuos speech recognition using the n-best algorithm. IEEE 2 (1990) 701-704
  8. Bridle, J.S.: Training stochastic model recognition algorithms as networks can lead to maximum mutual information estimation of parameters. Advances in neural information processing systems 2 (1990) 211-217
  9. B. H. Juang, L.R.R.: Hidden markov models for speech recognition. Technometrics 33 (1991) 251-272
  10. Lippi, V.: Design and development of a human gesture recognition system in threedimensional interactive virtual environment. Master's thesis, University of Pisa, College of engineering. (2008) Consultation Allowed.
  11. Stuart Russel, P.N.: Artificial Intelligence: A Modern Approach. Pearson (2003)
  12. Rodriguez, O.P., Avizzano, C., Sotgiu, E., Pabon, S., Frisoli, A., Ortiz, J., Bergamasco, M.: A wireless bluetooth dataglove based on a novel goniometric sensors. Robot and Human interactive Communication, 2007. RO-MAN 2007. The 16th IEEE International Symposium on 1 (2007) 1185- 1190
Download


Paper Citation


in Harvard Style

Lippi V., Ruffaldi E., Avizzano C. and Bergamasco M. (2009). Recognition of Hand Gestures Tracked by a Dataglove: Exploiting Hidden Markov Models Discriminative Training and Environment Description to Improve Recognition Performance . In Proceedings of the 5th International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: Workshop ANNIIP, (ICINCO 2009) ISBN 978-989-674-002-3, pages 33-42. DOI: 10.5220/0002206800330042


in Bibtex Style

@conference{workshop anniip09,
author={Vittorio Lippi and Emanuele Ruffaldi and Carlo Alberto Avizzano and Massimo Bergamasco},
title={Recognition of Hand Gestures Tracked by a Dataglove: Exploiting Hidden Markov Models Discriminative Training and Environment Description to Improve Recognition Performance},
booktitle={Proceedings of the 5th International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: Workshop ANNIIP, (ICINCO 2009)},
year={2009},
pages={33-42},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002206800330042},
isbn={978-989-674-002-3},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 5th International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: Workshop ANNIIP, (ICINCO 2009)
TI - Recognition of Hand Gestures Tracked by a Dataglove: Exploiting Hidden Markov Models Discriminative Training and Environment Description to Improve Recognition Performance
SN - 978-989-674-002-3
AU - Lippi V.
AU - Ruffaldi E.
AU - Avizzano C.
AU - Bergamasco M.
PY - 2009
SP - 33
EP - 42
DO - 10.5220/0002206800330042