Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations.
Thomas VerelstPaul K. RubensteinMarcin EichnerTinne TuytelaarsMaxim BermanPublished in: WACV (2023)
Keyphrases
- multi label
- multi label classification
- image annotation
- classifier training
- class labels
- multilabel classification
- text categorization
- training set
- image classification
- multi label learning
- binary classification
- multiple labels
- graph cuts
- label space
- text classification
- labeled images
- training examples
- multi instance
- training samples
- max margin
- training data
- automatic image annotation
- supervised learning
- multi label image annotation
- text classifiers
- classification algorithm
- naive bayes
- binary classifiers
- label assignment
- hierarchical text categorization
- decision trees
- support vector
- multi class
- feature selection
- label propagation
- training dataset
- visual features
- semi supervised learning
- feature extraction
- data samples
- visual concepts