Chor-rn generates new dance moves based on the style of a human choreographer. The system is trained on dance moves of a particular choreographer. To use the system to generate new moves, the choreographer choreographs a dance sequence, and Chor-rn produces a new sequence in response.
‘At the core of chor-rnn is a deep recurrent neural network trained on raw motion capture data and that can generate new dance sequences for a solo dancer.’