Semi-supervised metric learning from synchronized multi-view video, combined with a canonical yaw-aligned pose representation, improves monocular 3D pose estimation when 3D labels are scarce.
Position referencing and consistent world modeling for mobile robots
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimation
Semi-supervised metric learning from synchronized multi-view video, combined with a canonical yaw-aligned pose representation, improves monocular 3D pose estimation when 3D labels are scarce.