No abstract

This paper presents an attempt to generate novel dance movement based on motion captured human dance. Captured movements are analyzed statistically using nonlinear principal component analysis in order to create a "map" of observed poses. A method for automatically exploring the map by generating random trajectories is then presented, constituting a kind of improvisation. The result is an animated avatar that exhibits creative and novel movements in the style of its teacher, and paves the way towards a fully improvised live performance between an avatar and a human dancer.

No abstract

The paper describes an interaction between a human dancer and an improvising avatar, where the dancer gives kinetic feedback to the software in real time. By tracking the dancer's movements with a motion-capture camera and extracting basic motion features, the system detects feedback signals and lets them guide the avatar's behaviour. High intensity of movement by the dancer encourages novel and expansive behaviour in the avatar. Despite the crudeness and simplicity of the proposed mechanism, the high degree of feedback in both directions is expected to yield unpredictable and complex results. In comparison with more controlled settings, the open-endedness and complexity of this kinetic "dialogue" is likely to increase the creative potential of the exchange between dancer and software.