Spatio-Temporal Vector of Locally Max Pooled Features for Action Recognition in Videos.
Ionut Cosmin DutaBogdan IonescuKiyoharu AizawaNicu SebePublished in: CVPR (2017)
Keyphrases
- spatio temporal
- action recognition in videos
- feature vectors
- action recognition
- low level
- feature set
- prior knowledge
- co occurrence
- multimedia
- video representation
- spatial and temporal
- image sequences
- feature extraction
- image features
- classification accuracy
- face images
- machine learning
- text categorization
- wireless sensor networks
- data streams
- feature values
- feature space
- moving objects