Space-time Mixing Attention for Video Transformer.
Adrian BulatJuan-Manuel Perez-RuaSwathikiran SudhakaranBrais MartínezGeorgios TzimiropoulosPublished in: CoRR (2021)
Keyphrases
- space time
- video sequences
- spatial and temporal
- video representation
- spatio temporal
- input video
- dynamic scenes
- video data
- motion patterns
- human actions
- video content
- interesting events
- multiple view geometry
- video images
- video database
- video analysis
- machine learning
- visual attention
- moving objects
- three dimensional
- multimedia