Abstract | ||
---|---|---|
Realistic full-body avatar representation inside Virtual Reality is a big shortcoming of state-of-the-art VR systems. It remains a technically challenging task to capture human motion precisely without marker-based full-body tracking systems, which are expensive and impractical. Trying to tackle this challenge, we propose a simple yet efficient approach for avatar motion reconstruction. VIRTOOAIR (VIrtual Reality TOOlbox for Avatar Intelligent Reconstruction) combines Deep Learning for upper body reconstruction and most recent methods for single camera based pose recovery for the lower body parts. Our preliminary results demonstrate the advantages of our system's avatar pose reconstruction. This is mainly determined by the use of a powerful learning system, which offers significantly better results than existing heuristic solutions for inverse kinematics. Our system supports the paradigm shift towards learning systems capable to track full-body avatars inside Virtual Reality without the need of expensive external tracking hardware. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/ISMAR-Adjunct.2018.00085 | 2018 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) |
Keywords | Field | DocType |
Avatars,Cameras,Image reconstruction,Kinematics,Tracking,Skeleton,Quaternions | Iterative reconstruction,Computer vision,Heuristic,Virtual reality,Inverse kinematics,Computer science,Toolbox,Tracking system,Artificial intelligence,Deep learning,Avatar | Conference |
ISBN | Citations | PageRank |
978-1-5386-7592-2 | 0 | 0.34 |
References | Authors | |
0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Armin Becher | 1 | 0 | 1.35 |
Cristian Axenie | 2 | 10 | 5.25 |
Thomas Grauschopf | 3 | 0 | 1.35 |