POPE: 6-DoF Promptable Pose Estimation of Any Object, in Any Scene, with One Reference.
Zhiwen FanPanwang PanPeihao WangYifan JiangDejia XuHanwen JiangZhangyang WangPublished in: CoRR (2023)
Keyphrases
- pose estimation
- appearance information
- d objects
- camera views
- object recognition and pose estimation
- object pose
- depth images
- position and orientation
- degrees of freedom
- camera pose estimation
- pose parameters
- multiple objects
- multiple cameras
- invariant recognition
- computer vision
- human body
- rigid body motion
- camera pose
- ground plane
- human pose
- complex scenes
- moving objects
- feature points
- d scene
- object models
- implicit polynomial
- visual hull
- hand pose
- body parts
- three dimensional
- target object
- human pose estimation
- complex objects
- rigid objects
- video sequences
- object recognition
- dynamic scenes
- object appearance
- viewpoint
- multi view
- object segmentation
- object classes
- image sequences
- pose estimates
- single image
- structure from motion