Repaint: Improving the Generalization of Down-Stream Visual Tasks by Generating Multiple Instances of Training Examples.
Amin Banitalebi-DehkordiYong ZhangPublished in: CoRR (2021)
Keyphrases
- training examples
- visual tasks
- small number
- active learning
- learning algorithm
- training set
- training data
- supervised learning
- active vision
- object categories
- unlabeled data
- domain theory
- inductive bias
- labeled examples
- training samples
- labeled data
- image features
- data streams
- semi supervised learning
- mobile robot
- control system
- real time