Learning hierarchical spatio-temporal pattern for human activity prediction
详细信息 查看全文
- 作者：Wenwen Ding dww2048@163.com" class="auth_mail" title="E-mail the corresponding author ; Kai Liu ; kailiu@mail.xidian.edu.cn" class="auth_mail" title="E-mail the corresponding author ; Fei Cheng chengfei8582@163.com" class="auth_mail" title="E-mail the corresponding author ; Jin Zhang jinzhang.cv@gmail.com" class="auth_mail" title="E-mail the corresponding author
- 关键词：Skeleton joints ; 3D action feature representation ; Self-organizing map ; Hebbian learning ; Variable order Markov model ; Probabilistic suffix tree ; RGB-D dataset ; 3D trajectory segmentation
- 刊名：Journal of Visual Communication and Image Representation
- 出版年：2016
- 出版时间：February 2016
- 年：2016
- 卷：35
- 期：Complete
- 页码：103-111
- 全文大小：1226 K
文摘
- •
A novel approach to learning a hierarchical spatio-temporal pattern of human actions.
- •
Spatio-temporal pattern can be learned by a Hierarchical Self-Organizing Map (HSOM).
- •
The associative weights between HSOM can be obtained through Hebbian learning.
- •
Ongoing activities can be predicted by Variable order Markov Model (VMM).