Improving Visual Question Answering by Semantic Segmentation.
Viet-Quoc PhamNao MishimaToshiaki NakasuPublished in: ICANN (3) (2021)
Keyphrases
- question answering
- semantic segmentation
- weakly supervised
- conditional random fields
- relation extraction
- information retrieval
- information extraction
- natural language processing
- superpixels
- named entities
- natural language
- object categories
- scene classification
- visual information
- image understanding
- visual features
- machine learning
- object classes
- object detection
- artificial intelligence
- multi modal
- image set
- graphical models
- named entity recognition
- low level
- object recognition
- multiscale
- learning algorithm