2-D Skeleton-Based Action Recognition via Two-Branch Stacked LSTM-RNNs.
Danilo AvolaMarco CascioLuigi CinqueGian Luca ForestiCristiano MassaroniEmanuele RodolàPublished in: IEEE Trans. Multim. (2020)
Keyphrases
- action recognition
- recurrent neural networks
- long short term memory
- human actions
- bag of words
- neural network
- activity recognition
- computer vision
- human detection
- feed forward
- spatial temporal
- body parts
- bag of features
- action classification
- human activities
- mid level
- action primitives
- view invariant
- recognizing actions
- static images
- action detection
- artificial neural networks
- recognition of human actions
- motion features
- motion recognition
- feature extraction
- recognizing human actions
- action recognition in videos
- independent subspace analysis
- space time interest points
- machine learning