GazeVQA: A Video Question Answering Dataset for Multiview Eye-Gaze Task-Oriented Collaborations.
Muhammet IlaslanChenan SongJoya ChenDifei GaoWeixian LeiQianli XuJoo LimMike Zheng ShouPublished in: EMNLP (2023)
Keyphrases
- question answering
- eye gaze
- video teleconferencing
- eye contact
- eye tracking
- information retrieval
- information extraction
- eye movements
- natural language
- question classification
- natural language questions
- video content
- human actions
- video streams
- natural language processing
- question answering systems
- open domain question answering
- qa clef
- passage retrieval
- multimedia
- video frames
- video sequences
- syntactic information
- video data
- gaze direction
- cross language
- eye tracker
- multi view
- key frames
- answer extraction
- image retrieval
- answer validation
- speech transcripts
- multiple views