Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering.
Peter AndersonXiaodong HeChris BuehlerDamien TeneyMark JohnsonStephen GouldLei ZhangPublished in: CVPR (2018)
Keyphrases
- question answering
- low level
- named entities
- image features
- question classification
- passage retrieval
- visual data
- cross language
- natural language processing
- image classification
- information retrieval
- question answering systems
- information extraction
- image retrieval
- image representation
- relation extraction
- open domain question answering
- visual features
- expert systems
- data mining
- sentence retrieval
- natural language questions
- wordnet
- image content
- natural language
- artificial intelligence
- qa clef
- answer validation