.png){kind=icon}
返回
Spatial-Temporal Dynamic Graph Attention Network for Skeleton-Based Action Recognition
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2023.3247820.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Human body skeleton, acting as a spatiotemporal graph, is increasing attentions of researchers to adopt graph convolutional networks (GCN) to mine the discriminative features from skeleton joints. However, one of GCN's flaws is its inability to handle long-distance reliance between joints. In this regard, graph attention network (GAT) was recently suggested, which combines graph convolutions with a self-attention mechanism to extract the most informative joint of a human skeleton and increase the model accuracy. However, GAT can compute only static attention: for each query node, the attention rank is same which severely hurts and limits the expressivity of an attention mechanism. In this work, we present a spatial-temporal dynamic graph attention network (ST-DGAT) to learn the spatial-temporal patterns of skeleton sequences. For dynamic graph attention, we tweak the order of weighted vector operations in GAT, our approach achieves a global approximate attention function, making it strictly superior to GAT. Experiments show that by fixing the order of internal operation of GAT the proposed model achieved better action classification results while maintaining the same computing cost as GAT. The proposed framework has been evaluated on well-known publicly available large-scale datasets NTU60, NTU120, and Kinetics-400, which notably outperforms state-of-the-art (SOTA) results with an accuracy of 96.4%, 88.2%, and 61.0%, respectively.
Keyword:
Computational modeling
Spatiotemporal phenomena
Feature extraction
Convolutional neural networks
Computer architecture
Skeleton
action recognition
graph attention network
multihead attention
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
引用论文
Suspicious Activity Recognition Using Proposed Deep L4-Branched-Actionnet With Entropy Coded Ant Colony System Optimization基于熵编码蚁群系统优化的深度L4-Branched-Actionnet可疑活动识别.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏Evaluation of the Effect of Systolic Blood Pressure and Pulse Pressure on Cognitive Function: The Women's Health and Aging Study IIPLoS ONEOAAI
PLoS ONEIF0
OAAI分享收藏Attention adjacency matrix based graph convolutional networks for skeleton-based action recognition基于注意邻接矩阵的图卷积网络在骨架动作识别中的应用NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏
查看更多内容