Recurrent Graph Convolutional Autoencoder for Unsupervised Skeleton-Based Action Recognition.
Han YaoS.-J. ZhaoChi XieKenan YeShuang LiangPublished in: ICME (2021)
Keyphrases
- action recognition
- restricted boltzmann machine
- independent subspace analysis
- human actions
- bag of words
- activity recognition
- human detection
- computer vision
- body parts
- action classification
- bag of features
- recognizing human actions
- deep learning
- recognizing actions
- view invariant
- video dataset
- human pose
- spatial temporal
- graph structure
- recognition of human actions
- semi supervised
- weakly supervised
- human activities
- supervised learning
- action primitives
- view invariant action recognition
- independent component analysis
- conditional random fields
- image features