Two-Stream LSTM for Action Recognition with RGB-D-Based Hand-Crafted Features and Feature Combination.
Yun HanSheng-Luen ChungSheng-Fang ChenShun-Feng SuPublished in: SMC (2018)
Keyphrases
- action recognition
- feature set
- hand crafted
- feature vectors
- depth cameras
- image features
- human actions
- bag of words
- computer vision
- feature extraction
- feature space
- linguistic features
- classification accuracy
- low level
- feature descriptors
- data streams
- co occurrence
- domain independent
- structural features
- learning algorithm