• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on social robotics >An Attention-Aware Model for Human Action Recognition on Tree-Based Skeleton Sequences
【24h】

An Attention-Aware Model for Human Action Recognition on Tree-Based Skeleton Sequences

机译：基于树的骨架序列上的人类动作识别注意模型

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Skeleton-based human action recognition (HAR) has attracted a lot of research attentions because of robustness to variations of locations and appearances. However, most existing methods treat the whole skeleton as a fixed pattern, in which the importance of different skeleton joints for action recognition is not considered. In this paper, a novel CNN-based attention-ware network is proposed. First, to describe the semantic meaning of skeletons and learn the discriminative joints over time, an attention generate network named Global Attention Network (GAN) is proposed to generate attention masks. Then, to encode the spatial structure of skeleton sequences, we design a tree-based traversal (TTTM) rule, which can represent the skeleton structure, as a convolution unit of main network. Finally, the GAN and main network are cascaded as a whole network which is trained in an end-to-end manner. Experiments show that the TTTM and GAN are supplemented each other, and the whole network achieves an efficient improvement over the state-of-the-arts, e.g., the classification accuracy of this network was 83.6% and 89.5% on NTU-RGBD CV and CS dataset, which outperforms any other methods.
机译：基于骨架的人类动作识别（HAR）由于对位置和外观的变化具有鲁棒性，因此吸引了许多研究关注。然而，大多数现有方法将整个骨骼视为固定模式，其中没有考虑不同骨骼关节对于动作识别的重要性。本文提出了一种基于CNN的新型注意力软件网络。首先，为了描述骨骼的语义含义并随着时间学习判别关节，提出了一种名为全球注意力网络（GAN）的注意力生成网络来生成注意力掩码。然后，为了编码骨架序列的空间结构，我们设计了一个可以表示骨架结构的基于树的遍历（TTTM）规则作为主网络的卷积单元。最后，GAN和主网络被级联为一个完整的网络，并以端到端的方式进行训练。实验表明，TTTM和GAN相辅相成，整个网络实现了对现有技术的有效改进，例如，该网络在NTU-RGBD CV和CTU上的分类精度分别为83.6％和89.5％。 CS数据集，其性能优于任何其他方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号