Reinforcement Learning for Picking Cluttered General Objects with Dense Object Descriptors.
Hoang-Giang CaoWeihao ZengI-Chen WuPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- object model
- d objects
- object representation
- multiple objects
- keypoints
- moving objects
- data objects
- similar objects
- partially occluded objects
- visual appearance
- individual objects
- digital objects
- visual objects
- target object
- multi object
- spatial relationships
- spatial relations
- objects represented
- complex scenes
- real objects
- object description
- object position
- bounding box
- object segmentation
- object classes
- object level
- complex objects
- object representations
- real world objects
- object identification
- special case
- cluttered images
- mobile objects
- shape descriptors
- object shape
- partial occlusion
- articulated objects
- object models
- function approximation
- database objects
- rigid objects
- object features
- deformable objects
- object localization
- markov decision processes
- background clutter
- image regions
- object detection