DDGCN: A Dynamic Directed Graph Convolutional Network for Action Recognition
Introduction 作者认为解决如下两个问题能有效增强GCN在动作识别中的能力： 1.在人类骨骼的不同部位中有着时空关联性，但这些关联性是动态的，而且在时空域中不同的动作关联性也是不同的。因此提取这些关联性很困难，标椎卷积操作普遍采纳的传统GCN是静态的，而且仅仅描述了邻居节点的空间联系，因而不能准确的获得这样的动态时空联系。 2.骨骼的空间层次结构和运动的时间序列特性都编码了顺序信息，这在
相关文章
- 1. paper reading：Part-based Graph Convolutional Network for Action Recognition
- 2. Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
- 3. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
- 4. 【论文翻译】An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
- 5. An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
- 6. Part-based Graph Convolutional Network for Action Recognition论文笔记
- 7. 译：Convolutional Two-Stream Network Fusion for Video Action Recognition
- 8. [NLU] DialogueGCN: A Graph Convolutional Neural Network for Emotion Recognition in Conversation
- 9. Emotion Recognition Using Graph Convolutional Networks
- 10. Graph Convolutional Networks for Temporal Action Localization
- 更多相关文章...
- • Scala for循环 - Scala教程
- • Lua for 循环 - Lua 教程
- • IntelliJ IDEA代码格式化设置
- • Flink 数据传输及反压详解
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. paper reading：Part-based Graph Convolutional Network for Action Recognition
- 2. Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
- 3. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
- 4. 【论文翻译】An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
- 5. An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
- 6. Part-based Graph Convolutional Network for Action Recognition论文笔记
- 7. 译：Convolutional Two-Stream Network Fusion for Video Action Recognition
- 8. [NLU] DialogueGCN: A Graph Convolutional Neural Network for Emotion Recognition in Conversation
- 9. Emotion Recognition Using Graph Convolutional Networks
- 10. Graph Convolutional Networks for Temporal Action Localization