Pretraining boosts out-of-domain robustness for pose estimation.
Alexander MathisMert YüksekgönülByron RogersMatthias BethgeMackenzie W. MathisPublished in: CoRR (2019)
Keyphrases
- pose estimation
- d objects
- human body
- human pose estimation
- computer vision
- problems in computer vision
- feature points
- pose parameters
- degrees of freedom
- body parts
- position and orientation
- hand pose
- human pose
- pose recovery
- object pose
- depth images
- rotation and translation
- monocular images
- training set
- image based localization
- hypothesis verification
- object recognition and pose estimation
- articulated model
- pose estimates
- head pose estimation
- three dimensional