Improving Object Pose Estimation by Fusion With a Multimodal Prior - Utilizing Uncertainty-Based CNN Pipelines for Robotics.
Jesse Richter-KlugPatrick ManiaGayane KazhoyanMichael BeetzUdo FresePublished in: IEEE Robotics Autom. Lett. (2022)
Keyphrases
- pose estimation
- d objects
- object recognition and pose estimation
- computer vision
- object pose
- position and orientation
- pose recovery
- pose parameters
- invariant recognition
- degrees of freedom
- human body
- human pose
- human pose estimation
- pose determination
- body parts
- visual hull
- shape from silhouettes
- feature points
- monocular images
- multi view
- machine learning
- head pose estimation
- depth images
- implicit polynomial
- hand pose
- viewpoint
- object recognition
- upper body
- three dimensional
- articulated objects