Temporally Guided Music-to-Body-Movement Generation

This paper presents a neural network model to generate virtual violinistâĂŹs 3-D skeleton movements from music audio. Improved from the conventional recurrent neural network models for generating 2-D skeleton data in previous works, the proposed model incorporates an encoder-decoder architecture, as well as the selfattention mechanism to model the complicated dynamics in body movement sequences. To facilitate the optimization of self-attention model, beat tracking is applied to determine effective sizes and boundaries of the training examples. The decoder is accompanied with a refining network and a bowing attack inference mechanism to emphasize the right-hand behavior and bowing attack timing. Both objective and subjective evaluations reveal that the proposed model outperforms the state-of-the-art methods. To the best of our knowledge, this work represents the first attempt to generate 3-D violinistsâĂŹ body movements considering key features in musical body movement.

• Computing methodologies → Motion processing; • Applied computing → Media arts; Sound and music computing; • Humancentered computing → Sound-based input / output.