Movement sequence analysis using hidden Markov models: a case study in Tai Chi performance

Abstract

Movement sequences are essential to dance and expressive movement practice; yet, they remain underexplored in movement and computing research, where the focus on short gestures prevails. We propose a method for movement sequence analysis based on motion trajectory synthesis with Hidden Markov Models. The method uses Hidden Markov Regression for jointly synthesizing motion feature trajectories and their associated variances, that serves as basis for investigating performers' consistency across executions of a movement sequence. We illustrate the method with a use-case in Tai Chi performance, and we further extend the approach to cross-modal analysis of vocalized movements.