Full-Body Motion from a Single Head-Mounted Device: Generating SMPL Poses from Partial Observations
Authors:
- 1 October 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 11667-11677
Abstract
The increased availability and maturity of head-mounted and wearable devices opens up opportunities for remote communication and collaboration. However, the signal streams provided by these devices (e.g., head pose, hand pose, and gaze direction) do not represent a whole person. One of the main open problems is therefore how to leverage these signals to build faithful representations of the user. In this paper, we propose a method based on variational autoencoders to generate articulated poses of a human skeleton based on noisy streams of head and hand pose. Our approach relies on a model of pose likelihood that is novel and theoretically well-grounded. We demonstrate on publicly available datasets that our method is effective even from very impoverished signals and investigate how pose prediction can be made more accurate and realistic.Keywords
This publication has 54 references indexed in Scilit:
- SMPLACM Transactions on Graphics, 2015
- The KIT whole-body human motion database2015
- First-person pose recognition using egocentric workspaces2015
- Pose-conditioned joint angle limits for 3D human pose reconstruction2015
- Egocentric articulated pose tracking for action recognition2015
- Construction and optimal search of interpolated motion graphs2007
- SCAPE2005
- Kinematic jump processes for monocular 3D human tracking2003
- Interactive motion generation from examplesACM Transactions on Graphics, 2002
- Distance Metrics on the Rigid-Body Motions with Applications to Mechanism DesignJournal of Mechanical Design, 1995