Simple, Efficient and Effective Encodings of Local Deep Features for Video Action Recognition.
Ionut C. DutaBogdan IonescuKiyoharu AizawaNicu SebePublished in: ICMR (2017)
Keyphrases
- action recognition
- spatio temporal interest points
- human actions
- action classification
- motion features
- space time interest points
- static images
- spatial temporal
- bag of words
- mid level
- action detection
- human detection
- recognizing human actions
- bag of features
- activity recognition
- human activities
- recognition of human actions
- video dataset
- depth cameras
- video clips
- body parts
- video images
- computer vision
- motion history images
- multimedia
- feature vectors
- video data
- low level
- image features
- feature space
- global features
- temporal structure
- video sequences
- key frames
- event detection
- recognizing actions
- video frames
- feature extraction