What Do I Annotate Next? An Empirical Study of Active Learning for Action Localization.
Fabian Caba HeilbronJoon-Young LeeHailin JinBernard GhanemPublished in: ECCV (11) (2018)
Keyphrases
- active learning
- semi supervised
- selective sampling
- training examples
- machine learning
- training set
- random sampling
- labeled data
- semantic annotation
- pool based active learning
- batch mode active learning
- localization error
- learning strategies
- supervised learning
- relevance feedback
- action recognition
- human actions
- object localization
- batch mode
- localization algorithm
- source localization
- learning environment
- accurate localization
- active learning strategies
- decision trees
- information retrieval