LoL-V2T: Large-Scale Esports Video Description Dataset.
Tsunehiko TanakaEdgar Simo-SerraPublished in: CVPR Workshops (2021)
Keyphrases
- trecvid multimedia event detection
- event detection
- event recognition
- human actions
- video data
- video event detection
- real time
- multimedia
- weakly labeled
- video streams
- small scale
- video dataset
- video sequences
- high level
- content description
- real time video
- video analysis
- video frames
- video content
- symbolic descriptions
- video annotation
- real world
- key frames
- benchmark datasets
- feature set
- spatio temporal
- computer vision
- dynamic scenes
- video database
- digital video
- video images
- video clips
- video retrieval
- temporal information
- online video
- million images
- web videos
- space time