STAR++: Rethinking spatio-temporal cross attention transformer for video action recognition.
Dasom AhnSangwon KimByoungChul KoPublished in: Appl. Intell. (2023)
Keyphrases
- action recognition
- human actions
- spatio temporal interest points
- spatial temporal
- spatio temporal
- action classification
- recognition of human actions
- video database
- activity recognition
- video dataset
- action detection
- action recognition in videos
- view invariant
- human activities
- computer vision
- recognizing human actions
- bag of words
- human detection
- human motion
- mid level
- space time
- static images
- motion features
- spatial and temporal
- video sequences
- recognizing actions
- body parts
- moving objects
- image sequences
- machine learning
- space time interest points
- motion capture data
- bag of features
- motion history images
- depth sensors
- human pose
- motion patterns