Generating Diverse and Accurate Visual Captions by Comparative Adversarial Learning.
Dianqi LiQiuyuan HuangXiaodong HeLei ZhangMing-Ting SunPublished in: CoRR (2018)
Keyphrases
- learning process
- learning algorithm
- learning systems
- information retrieval
- visual features
- prior knowledge
- wide variety
- visual learning
- data sets
- unsupervised learning
- real world
- supervised learning
- online learning
- active learning
- high quality
- computationally efficient
- high level
- learning problems
- incremental learning