Weakly Supervised Annotations for Multi-modal Greeting Cards Dataset.
Sidra HanifLongin Jan LateckiPublished in: CoRR (2022)
Keyphrases
- multi modal
- weakly supervised
- image annotation
- object detectors
- relation extraction
- object class
- superpixels
- semi supervised
- topic models
- object detection
- high dimensional
- metadata
- named entities
- question answering
- natural images
- object categories
- automatic extraction
- information retrieval
- domain specific
- information extraction
- viewpoint
- object recognition
- video sequences
- keywords
- computer vision
- machine learning