DM-CTSA: a discriminative multi-focused and complementary temporal/spatial attention framework for action recognition.
Ming TongKaibo YanLei JinXing YueMingyang LiPublished in: Neural Comput. Appl. (2021)
Keyphrases
- action recognition
- spatial temporal
- action classification
- spatio temporal
- spatial and temporal
- human actions
- space time
- bag of words
- human detection
- max margin
- mid level
- motion history images
- video dataset
- computer vision
- body parts
- bag of features
- spatial information
- feature extraction
- recognition of human actions
- action recognition in videos
- depth sensors
- view invariant
- activity recognition
- three dimensional