Automatic Choreography Generation with Convolutional Encoder-decoder Network

Computer Science

Automatic choreography generation is a challenging task because it often requires an understanding of two abstract concepts music and dance which are realized in the two different modalities, namely audio and video, respectively. In this paper, we propose a music-driven choreography generation system using an auto-regressive encoderdecoder network. To this end, we first collected a set of multimedia clips that include both music and corresponding dance motion. We then extract the joint coordinates of the dancer from video and the mel-spectrogram of music from audio and train our network using musicchoreography pairs as input. Finally, a novel dance motion is generated at the inference time when only music is given as an input. We performed a user study for a qualitative evaluation of the proposed method, and the results show that the proposed model is able to generate musically meaningful and natural dance movements given an unheard song. We also revealed through quantitative evaluation that the network has created a movement that correlates with the beat of music.