[ISEA2015] Paper: Michael Junokas, Kyungho Lee, Mohammad Amazadeh & Guy Garnett – Capturing and Recognizing Expressive Performance Gesture

Abstract (Short paper)

Keywords: Machine Learning, Expressive Performance Gesture, Expressive Movement Recognition.

A better understanding and control of expressive performance gesture potentially could have a large and disruptive impact on electronic media and movement performance practice. We use digitally captured positional data, features extracted from this positional data, and a variety of machine-learning algorithms, to improve the accuracy of recognizing expressive qualities of performance gestures, using concepts derived from Laban Movement Analysis (LMA). Through these methods, we seek to develop better human-computer interfaces, to expand expressive movement vocabularies, and to shift movement aesthetics, by empowering users to exploit their full performance capabilities.

  • Michael J. Junokas, Research Assistant, Illinois Informatics Institute, University of Illinois, Urbana-Champaign, USA. He develops innovative, multi-platform systems that have the ability to gather, interpret, process, and control signals in live artistic performance. Through the exploration of these systems, he hopes to create immersive technological environments artists can use for their own creative pursuits.  junokas.wordpress.com
  • Kyungho Lee, Mohammad Amanzadeh & Guy E. Garnett, Illinois Informatics Institute at the University of Illinois, Urbana-Champaign, USA

Full text (PDF)  p. 370-373