VLAD3: Encoding Dynamics of Deep Features for Action Recognition.
Yingwei LiWeixin LiVijay MahadevanNuno VasconcelosPublished in: CVPR (2016)
Keyphrases
- action recognition
- bag of words
- human detection
- bag of features
- motion features
- static images
- computer vision
- activity recognition
- human actions
- body parts
- mid level
- human movements
- action classification
- deformable part models
- depth sensors
- low level
- feature extraction
- video dataset
- recognition of human actions
- motion history images
- action detection
- action primitives
- depth cameras
- human pose
- spatial temporal
- object recognition
- image search
- three dimensional
- recognizing actions
- machine learning