Using hidden Markov models and dynamic size functions
  for gesture recognition

Andrea Sorrentino, Fabio Cuzzolin, and Ruggero Frezza
The 8th British Machine Vision Conference (BMVC'97)

Automatic gesture recognition is an important and challenging problem in comupter vision. In this paper we present an original technique for hand gesture recognition based on a dynamic shape representation by combining size functions and hidden Markov models (HMM). Size functions are objects of topological nature which allow to describe shape with completeness and a remarkable tolerance to noise. HMM allow, instead, for the inclusion of dynamics in the model. Each gesture is described by a different probabilistic finite state machine which models a succession of so called canonical postures of the hand. The state dynamics describe the transition between canonical postures while the observation equations are maps from the set of canonical postures to size functions. Tests on real image sequences are included.
PDF  Zipped Postscript
BibTeX Entry

  AUTHOR = {A Sorrentino and F. Cuzzolin and R. Frezza}, 
  TITLE = {Using Hidden Markov Models and Dynamic Size Functions for Gesture Recognition}, 
  JOURNAL = {Proceedings of the British Machine Vision Conference (BMVC97)}, 
  VOLUME = {2},
  PAGES = {560--570},
  YEAR = {1997} 

INRIA Rhone-Alpes