Researching Mocap Solution for Virtual Real-Time Dance Duets

Updated: Apr 22

Future thinking Goldsmiths University team collaborates with Target3D to investigate how computational arts could use motion capture to bring the world closer together for new dance performances.

Dan Strutt, Lecturer in the Department of Media, Communications and Cultural Studies at Goldsmiths, University of London, and his team of postgraduate students and industry professionals have been researching the potential application of new forms of accessible wireless and markerless motion capture in the creation, rehearsal, teaching and performance of choreographic dance work.


Having only used mocap that was pre-captured and pre-rendered on the team’s previous dance projects, live, real-time and generative use of motion capture was always their goal...


The aim To connect two dancers in two different locations - one in London and one in Singapore - through the internet, to dance together within a virtual environment.


For Dan, this wasn’t simply about hyping the tech up to be the next big thing, instead by experimenting and exploring he wanted to probe beyond superficial interest to see how genuinely useful this would be for dancers as a meaningful tool for choreographic communication.


The solution

Supported by Target3D and working within our Hoxton Studio, the Goldsmiths team created a virtual environment to connect Mavin Khoo, Creative Associate of Akram Khan Dance Company, with Melissa Kwek from LASALLE College of the Arts. Both Khoo, in our London space, and Kwek from a Singapore studio, danced in Noitom's Perception Neuron 32 V2 mocap suits that captured their data and distributed it 6750 miles through the network in real time. Each suit's IMU sensors were strapped to different parts of the dancers bodies, allowing the dancers to work together to devise an aesthetics of movement specific to this mode of remote working. Data was streamed from the sensors, across the world, and into graphics software to be instantly rendered, such that Goldsmiths and Lasalle could see and respond to each other's movement. A series of generative visualisations of the dance data were projected onto the walls of both locations simultaneously, the dancers responded in their own spaces, and t