Unsupervised Learning of Human Action Categories in Still Images with Deep Representations.
Yunpeng ZhengXuelong LiXiaoqiang LuPublished in: ACM Trans. Multim. Comput. Commun. Appl. (2020)
Keyphrases
- unsupervised learning
- human actions
- object recognition
- three dimensional
- image features
- image retrieval
- image classification
- temporal sequences
- spatio temporal
- action recognition
- deep architectures
- deep learning
- visual features
- feature points
- image representation
- supervised learning
- activity recognition
- human motion
- video sequences
- video images
- feature extraction
- high level
- machine learning