Motion-Appearance Co-Memory Networks for Video Question Answering.
Jiyang GaoRunzhou GeKan ChenRam NevatiaPublished in: CoRR (2018)
Keyphrases
- relevance feedback
- question answering
- object motion
- information retrieval
- dynamic textures
- space time
- video sequences
- image sequences
- key frames
- question classification
- information retrieval systems
- image retrieval
- passage retrieval
- video data
- natural language
- natural language processing
- information extraction
- syntactic information
- cross language
- multimedia
- qa clef
- video content
- video frames
- semantic roles
- visual data
- human motion
- natural language questions
- open domain question answering
- video retrieval
- question answering systems
- named entities
- qa systems
- knowledge representation
- answering questions
- answer validation