Sequential inter-hop graph convolution neural network (SIhGCN) for skeleton-based human action recognition.
Feri SetiawanBernardo Nugroho YahyaSeok-Ju ChunSeok-Lyong LeePublished in: Expert Syst. Appl. (2022)
Keyphrases
- action recognition
- neural network
- static images
- human movements
- human actions
- bag of words
- human activities
- activity recognition
- human detection
- motion capture data
- motion history images
- human object interactions
- computer vision
- action classification
- spatio temporal interest points
- body parts
- spatial temporal
- recognition of human actions
- recognizing human actions
- bag of features
- view invariant
- depth sensors
- graph structure
- directed graph
- action primitives
- video dataset
- action recognition in videos
- recognizing actions
- view invariant action recognition
- independent subspace analysis
- feature extraction
- multiscale
- spatio temporal
- human pose