Higher-Order Recurrent Network with Space-Time Attention for Video Early Action Recognition.
Tsung-Ming TaiGiuseppe FiameniCheng-Kuang LeeOswald LanzPublished in: ICIP (2022)
Keyphrases
- space time
- human actions
- action recognition
- higher order
- recurrent networks
- video sequences
- spatio temporal
- spatial and temporal
- action classification
- recognition of human actions
- recurrent neural networks
- dynamic scenes
- recognizing human actions
- bag of words
- activity recognition
- video dataset
- motion patterns
- video database
- computer vision
- action detection
- body parts
- neural network
- feed forward
- bag of features
- video shots
- mid level
- motion features
- multimedia
- human pose
- biologically inspired
- human activities
- video streams
- space time interest points