ResLNet: deep residual LSTM network with longer input for action recognition.
Tian WangJiakun LiHuai-Ning WuCe LiHichem SnoussiYang WuPublished in: Frontiers Comput. Sci. (2022)
Keyphrases
- action recognition
- human actions
- bag of words
- activity recognition
- human detection
- recognition of human actions
- action classification
- computer vision
- view invariant
- body parts
- spatial temporal
- mid level
- static images
- recurrent neural networks
- motion features
- video dataset
- wireless sensor networks
- action detection
- action primitives
- bag of features
- recognizing human actions
- view invariant action recognition
- recognizing actions
- independent subspace analysis
- depth sensors
- max margin
- visual features
- spatio temporal
- data streams
- three dimensional