TVQA+: Spatio-Temporal Grounding for Video Question Answering.
Jie LeiLicheng YuTamara L. BergMohit BansalPublished in: CoRR (2019)
Keyphrases
- question answering
- spatio temporal
- natural language
- named entities
- human actions
- information retrieval
- relation extraction
- multimedia
- qa clef
- natural language processing
- cross language
- information extraction
- video frames
- open domain question answering
- natural language questions
- video data
- video content
- question answering systems
- qa systems
- video sequences
- question classification
- passage retrieval
- syntactic information
- video database
- answer validation
- answer extraction
- answering questions
- semantic roles
- expert systems
- video shots
- candidate answers
- relational databases
- sentence retrieval
- key frames
- video retrieval