Semantic-Disentangled Transformer With Noun-Verb Embedding for Compositional Action Recognition.
Peng HuangRui YanXiangbo ShuZhewei TuGuangzhao DaiJinhui TangPublished in: IEEE Trans. Image Process. (2024)
Keyphrases
- action recognition
- natural language
- dependency relations
- noun phrases
- human actions
- bag of words
- spatial temporal
- activity recognition
- body parts
- computer vision
- human detection
- recognizing human actions
- recognition of human actions
- depth sensors
- recognizing actions
- bag of features
- human pose
- static images
- semantic relations
- view invariant
- high level
- action classification
- action detection
- action primitives
- spatio temporal
- action recognition in videos
- independent subspace analysis
- view invariant action recognition
- mid level
- human activities
- semantic information
- wordnet
- object recognition