Self-supervised Keypoint Correspondences for Multi-Person Pose Estimation and Tracking in Videos.
Umer RafiAndreas DoeringBastian LeibeJuergen GallPublished in: CoRR (2020)
Keyphrases
- pose estimation
- camera pose
- markerless
- pose recovery
- line correspondences
- human pose
- tracking and pose estimation
- monocular images
- pose parameters
- camera pose estimation
- degrees of freedom
- d objects
- pose estimates
- upper body
- human body
- feature points
- position and orientation
- human pose estimation
- computer vision
- image features
- articulated model
- video surveillance
- motion tracking
- point correspondences
- head pose estimation
- object tracking
- keypoints
- body parts
- static images
- moving camera
- focal length
- particle filtering
- motion segmentation
- kalman filter
- human motion
- video sequences
- object recognition and pose estimation
- mean shift
- viewpoint
- articulated objects
- depth images
- pictorial structure
- visual tracking
- video frames
- particle filter
- object detection
- face recognition