'Double Agent' is an interactive augmented environment, commissioned by the Museum of Discovery (MOD), Adelaide, and premiered as part of the MOD.ify exhibition, 2018. People physically interact with two projected virtual 'agents' within a large-scale three-dimensional projection. The agents are composed from hundreds of small invisible elements that are both drawn to and repelled by the movement of human bodies in the installation space, the relations between the elements visualised as complex tensile and fluid structures. One agent (the lower) interacts directly with people (interactors) in the installation space, building a physical model of the interactors bodies. The other agent (the higher) is generated from a simulated interactor which has learned how to dance and which improvises its dance moves in response to the human interactors' movements. Employing many hours of dance data, acquired through the motion capture of dancer's improvising within Double Agent, the simulated interactor agent has, through the use of machine-learning algorithms (Long Short-Term Memory Recurrent Neural Networks), developed the ability to improvise its own dance movements, responding to the presence of viewers who co-habit the installation space.
Artist and Systems Developer: Simon Biggs
Choreographer: Articulate Animal - Sue Hawksley and Tammy Arjona
Machine-Learning Engineering: Samya Bagchi
Scientific Advisor: Professor Mark McDonnell