CTM: Cross-time temporal module for fine-grained action recognition.
Huifang QianJialun ZhangJianping YiZhenyu ShiYimin ZhangPublished in: Comput. Vis. Image Underst. (2024)
Keyphrases
- fine grained
- action recognition
- spatial temporal
- coarse grained
- human actions
- motion history images
- human detection
- bag of words
- activity recognition
- action classification
- computer vision
- bag of features
- recognizing human actions
- spatio temporal
- body parts
- view invariant
- recognizing actions
- temporal reasoning
- video dataset
- human activities
- spatial and temporal
- space time
- data lineage
- temporal information
- human pose
- temporal relations
- machine learning
- action detection
- depth sensors
- access control
- sensor data
- graphical models
- image classification
- action primitives
- temporal structure
- visual words