Inferring Occluded Geometry Improves Performance when Retrieving an Object from Dense Clutter.
Andrew PriceLinyi JinDmitry BerensonPublished in: CoRR (2019)
Keyphrases
- object tracking
- partially occluded
- background clutter
- d objects
- object model
- geometric information
- moving objects
- information retrieval
- complex objects
- object shape
- computer vision
- occluded objects
- geometrical constraints
- shape reconstruction
- object segmentation
- three dimensional
- complex scenes
- planar objects
- single image
- geometric objects