TY - GEN
T1 - Hierarchical Graph Convolutional Network for Skeleton-Based Action Recognition
AU - Huang, Linjiang
AU - Huang, Yan
AU - Ouyang, Wanli
AU - Wang, Liang
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Skeleton-based action recognition has drawn much attention recently. Previous methods mainly focus on using RNNs or CNNs to process skeletons. But they ignore the topological structure of the skeleton which is very important for action recognition. Recently, Graph Convolutional Networks (GCNs) achieve remarkable performance in modeling non-Euclidean structures. However, current graph convolutional networks lack the capacity of modeling hierarchical information, which may be sub-optimal for classifying actions which are performed in a hierarchical way. In this work, a novel Hierarchical Graph Convolutional Network (HiGCN) is proposed to deal with these problems. The proposed model includes several Hierarchical Graph Convolutional Layers (HiGCLs). Each layer consists of an attention block and a hierarchical graph convolutional block, which are used for salient feature enhancement and hierarchical representation learning, respectively. To represent hierarchical information of human actions, we propose a graph pooling method, which is differentiable and can be plugged into GCN in an end-to-end manner. Extensive experiments on two benchmark datasets show the state-of-the-art performance of our method.
AB - Skeleton-based action recognition has drawn much attention recently. Previous methods mainly focus on using RNNs or CNNs to process skeletons. But they ignore the topological structure of the skeleton which is very important for action recognition. Recently, Graph Convolutional Networks (GCNs) achieve remarkable performance in modeling non-Euclidean structures. However, current graph convolutional networks lack the capacity of modeling hierarchical information, which may be sub-optimal for classifying actions which are performed in a hierarchical way. In this work, a novel Hierarchical Graph Convolutional Network (HiGCN) is proposed to deal with these problems. The proposed model includes several Hierarchical Graph Convolutional Layers (HiGCLs). Each layer consists of an attention block and a hierarchical graph convolutional block, which are used for salient feature enhancement and hierarchical representation learning, respectively. To represent hierarchical information of human actions, we propose a graph pooling method, which is differentiable and can be plugged into GCN in an end-to-end manner. Extensive experiments on two benchmark datasets show the state-of-the-art performance of our method.
KW - Action recognition
KW - Hierarchical graph convolutional network
KW - Skeleton
UR - https://www.scopus.com/pages/publications/85076910169
U2 - 10.1007/978-3-030-34120-6_8
DO - 10.1007/978-3-030-34120-6_8
M3 - 会议稿件
AN - SCOPUS:85076910169
SN - 9783030341190
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 93
EP - 102
BT - Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 1
A2 - Zhao, Yao
A2 - Lin, Chunyu
A2 - Barnes, Nick
A2 - Chen, Baoquan
A2 - Westermann, Rüdiger
A2 - Kong, Xiangwei
PB - Springer
T2 - 10th International Conference on Image and Graphics, ICIG 2019
Y2 - 23 August 2019 through 25 August 2019
ER -