Human Action Recognition Based on Sparse LSTM Auto-encoder and Improved 3D CNN.
Qi FuShiwei MaLina LiuJinjin LiuPublished in: ICNC-FSKD (2018)
Keyphrases
- action recognition
- static images
- human movements
- human actions
- bag of words
- activity recognition
- human activities
- spatio temporal interest points
- motion history images
- action classification
- human detection
- computer vision
- spatial temporal
- human object interactions
- recognizing human actions
- recognition of human actions
- body parts
- motion capture data
- view invariant
- video dataset
- human pose
- motion estimation
- action detection
- depth sensors
- independent subspace analysis
- space time interest points
- human activity recognition
- bag of features
- sparse representation
- low level
- face recognition