Visual Transformers with Primal Object Queries for Multi-Label Image Classification.
Vacit Oguz YaziciJoost van de WeijerLonglong YuPublished in: CoRR (2021)
Keyphrases
- multi label
- image classification
- visual features
- multi label classification
- data objects
- image annotation
- text categorization
- multi label learning
- image features
- binary classification
- bag of words
- visual information
- feature extraction
- image representation
- graph cuts
- visual words
- query language
- support vector
- visual concepts
- multiple labels
- text classification
- automatic image annotation
- max margin
- object level
- label assignment
- neural network
- scene classification
- visual similarity
- protein function prediction