Reinforcement Learning-Based Grasping via One-Shot Affordance Localization and Zero-Shot Contrastive Language-Image Learning.
Xiang LongLuke BeddowDenis HadjivelichkovAndromachi Maria DelfakiHelge WurdemannDimitrios KanoulasPublished in: SII (2024)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- input image
- image representation
- supervised learning
- image data
- image features
- function approximation
- image retrieval
- image analysis
- image segmentation
- multiscale
- mobile robot
- image classification
- object categories
- object oriented programming
- object localization
- pixel values
- robot control
- learning tasks
- markov decision processes
- image content
- spatial information
- single image
- image quality
- programming language
- object detection
- high resolution
- object recognition