Audio to Body Dynamics

Title: Audio to Body Dynamics

Teams: Facebook

Writers: Eli Shlizerman, Lucio Dery, Hayden Schoen, Ira Kemelmacher Shlizerman

Publication date: June 18, 2018

Abstract

We present a method that gets as input an audio of violin or piano playing, and outputs a video of skeleton predictions which are further used to animate an avatar. The key idea is to create an animation of an avatar that moves their hands similarly to how a pianist or violinist would do, just from audio. Notably, it’s not clear if body movement can be predicted from music at all and our aim in this work is to explore this possibility. In this paper, we present the first result that shows that natural body dynamics can be predicted. We built an LSTM network that is trained on violin and piano recital videos uploaded to the Internet. The predicted points are applied onto a rigged avatar to create the animation.