Reinforcement Learning for Picking Cluttered General Objects with Dense Object Descriptors.
Hoang-Giang CaoWeihao ZengI-Chen WuPublished in: ICRA (2022)
Keyphrases
- reinforcement learning
- multiple objects
- object representation
- object model
- d objects
- keypoints
- moving objects
- object identification
- data objects
- individual objects
- object level
- spatial relationships
- object features
- object segmentation
- visual objects
- visual appearance
- objects represented
- object position
- multi object
- similar objects
- target object
- object classes
- object representations
- object description
- spatial relations
- partial occlusion
- partially occluded objects
- background clutter
- object models
- bounding box
- real objects
- object localization
- object recognition
- cluttered images
- special case
- digital objects
- machine learning
- object appearance
- state space
- pose estimation
- object orientation
- real world objects
- appearance model
- rigid objects
- deformable objects