Gesture recognition in flow based on PCA analysis using multiagent system

  • Authors:
  • Ronan Billon;Alexis Nédélec;Jacques Tisseau

  • Affiliations:
  • Centre Européen de Réalité, Virtuelle, Plouzané, France;Centre Européen de Réalité, Virtuelle, Plouzané, France;Centre Européen de Réalité, Virtuelle, Plouzané, France

  • Venue:
  • ACE '08 Proceedings of the 2008 International Conference on Advances in Computer Entertainment Technology
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Our context is Virtual Theater. Our aim is to put on a short play featuring a real actor and a virtual actor, who will communicate through movements and choreography with mutual synchronization. We currently work on a system that can recognize key-gestures made by a real actor. In this paper, we cover a method for real-time recognition. Our idea is that the data for movements from any motion-capture system can be reduced to a single artificial signature. We use properties from Principal Component Analysis (PCA) to generate it. This artificial gesture representation is used in real-time by our multiagent systems to simultaneously perform segmentation (gesture's beginning and end) and recognition. We conducted various experiments which demonstrate our system and define its limitations.