Keyword-Aware Relative Spatio-Temporal Graph Networks for Video Question Answering.
Yi ChengHehe FanDongyun LinYing SunMohan S. KankanhalliJoo-Hwee LimPublished in: IEEE Trans. Multim. (2024)
Keyphrases
- question answering
- spatio temporal
- human actions
- information retrieval
- information extraction
- named entities
- natural language processing
- video content
- qa clef
- cross language
- video data
- natural language questions
- passage retrieval
- open domain question answering
- multimedia
- video sequences
- syntactic information
- image sequences
- keywords
- natural language
- video frames
- structured data
- question classification
- key frames
- semantic roles
- keyword search
- action recognition
- sentence retrieval
- question answering systems
- text summarization
- video retrieval
- candidate answers
- visual data
- answer extraction
- data mining
- graph structure
- document collections