Uncertainty-Aware Mapping from 3D Keypoints to Anatomical Landmarks for Markerless Biomechanics

1. 1Extract 3D skeletal keypoints from video
2. 2Estimate uncertainty of keypoints using predictive models
3. 3Map keypoints to 3D anatomical landmarks with uncertainty-aware approach
4. 4Use uncertainty measures for frame-wise quality control in downstream biomechanical mappings

This research benefits computer vision engineers and biomechanics researchers who work with 3D skeletal keypoints and anatomical landmarks, as it provides a principled mechanism for frame-wise quality control

💡 Predictive uncertainty can be used as a quantitative measure of confidence for mapping 3D pose keypoints to 3D anatomical landmarks

💡 Uncertainty-aware mapping for markerless biomechanics!