BehanceQA: A New Dataset for Identifying Question-Answer Pairs in Video Transcripts.
Amir Pouran Ben VeysehViet Dac LaiFranck DernoncourtThien Huu NguyenPublished in: LREC (2022)
Keyphrases
- multimedia
- human actions
- video data
- video sequences
- weakly labeled
- video frames
- video streams
- question answer pairs
- spatial and temporal
- video analysis
- database
- video content
- video database
- video clips
- event recognition
- video annotation
- space time
- question answering
- real time
- digital video
- trecvid multimedia event detection
- video retrieval
- benchmark datasets
- event detection
- human activities
- human generated
- dimensionality reduction
- artificial intelligence