GPM: A Generic Probabilistic Model to Recover Annotator's Behavior and Ground Truth Labeling.
Jing LiSuiyi LingJunle WangZhi LiPatrick Le CalletPublished in: CoRR (2020)
Keyphrases
- probabilistic model
- ground truth
- language model
- graphical models
- active learning
- high quality
- semantic annotation
- bayesian networks
- image segmentation
- databases
- image labeling
- quantitative evaluation
- information systems
- gold standard
- behavior patterns
- real time
- human behavior
- manually labeled
- conditional random fields
- semantic web
- unsupervised learning
- domain specific
- pairwise
- video sequences
- multi agent