Two-Stage Multimodality Fusion for High-Performance Text-Based Visual Question Answering.
Bingjia LiJie WangMinyi ZhaoShuigeng ZhouPublished in: ACCV (4) (2022)
Keyphrases
- question answering
- visual features
- information retrieval
- cross language
- natural language processing
- natural language
- visual information
- question answering systems
- named entities
- information extraction
- natural language questions
- question classification
- qa clef
- passage retrieval
- low level
- sentence retrieval
- open domain question answering
- answer validation
- syntactic information
- relation extraction
- answering questions
- automatically generated
- multimedia