Learning to Generate Grounded Visual Captions Without Localization Supervision.
Chih-Yao MaYannis KalantidisGhassan AlRegibPeter VajdaMarcus RohrbachZsolt KiraPublished in: ECCV (18) (2020)
Keyphrases
- active learning
- learning process
- learning algorithm
- reinforcement learning
- visual features
- learning problems
- online learning
- visual learning
- machine learning
- supervised learning
- visual processing
- inductive inference
- learning tasks
- visual information
- background knowledge
- learning systems
- data sets
- artificial neural networks
- object recognition
- video sequences
- neural network