Gaining Extra Supervision via Multi-task learning for Multi-Modal Video Question Answering.
Junyeong KimMinuk MaKyungsu KimSungjin KimChang D. YooPublished in: CoRR (2019)
Keyphrases
- multi modal
- question answering
- multi task learning
- multi task
- semantic concepts
- video search
- gaussian processes
- learning tasks
- video data
- information retrieval
- natural language processing
- video sequences
- transfer learning
- information extraction
- natural language
- high order
- video frames
- active learning
- high dimensional
- video content
- multimedia
- learning algorithm
- multimedia data
- image annotation
- video retrieval
- audio visual
- feature selection