This paper appeared at IEEE AI/VR 2020.
Mar Gonzalez-Franco, Zelia Egan, Matthew Peachey, Angus Antley, Tanmay Randhavane, Payod Panda, Yaying Zhang, Cheng Yao Wang, Derek F. Reilly, Tabitha C Peck, Andrea Stevenson Won, Anthony Steed and Eyal Ofek.
This paper presents MoveBox an open sourced toolbox for animating motion captured (MoCap) movements onto the Microsoft Rocketbox library of avatars. Motion capture is performed using a single depth sensor, such as Azure Kinect or Windows Kinect V2. Motion capture is performed in real-time using a single depth sensor, such as Azure Kinect or Windows Kinect V2, or extracted from existing RGB videos offline leveraging deep-learning computer vision techniques. Our toolbox enables real-time animation of the user’s avatar by converting the transformations between systems that have different joints and hierarchies. Additional features of the toolbox include recording, playback and looping animations, as well as basic audio lip sync, blinking and resizing of avatars as well as finger and hand animations. Our main contribution is both in the creation of this open source tool as well as the validation on different devices and discussion of MoveBox’s capabilities by end users.