GAS-GCN: Gated Action-Specific Graph Convolutional Networks for Skeleton-Based Action Recognition

Skeleton-based action recognition has achieved great advances with the development of graph convolutional networks (GCNs). Many existing GCNs-based models only use the fixed hand-crafted adjacency matrix to describe the connections between human body joints. This omits the important implicit connect...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Sensors Ročník 20; číslo 12; s. 3499
Hlavní autori: Chan, Wensong, Tian, Zhiqiang, Wu, Yang
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Switzerland MDPI AG 21.06.2020
MDPI
Predmet:
action recognition
Chemical technology
Datasets as Topic
Deep learning
gated convolutional neural networks
graph convolutional networks
Hands
Humans
Joints
Joints - physiology
Methods
Neural networks
Neural Networks, Computer
Skeleton
Skeleton - physiology
TP1-1185
deep learning
gated convolutional neural networks
graph convolutional networks
action recognition
ISSN:1424-8220, 1424-8220
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Skeleton-based action recognition has achieved great advances with the development of graph convolutional networks (GCNs). Many existing GCNs-based models only use the fixed hand-crafted adjacency matrix to describe the connections between human body joints. This omits the important implicit connections between joints, which contain discriminative information for different actions. In this paper, we propose an action-specific graph convolutional module, which is able to extract the implicit connections and properly balance them for each action. In addition, to filter out the useless and redundant information in the temporal dimension, we propose a simple yet effective operation named gated temporal convolution. These two major novelties ensure the superiority of our proposed method, as demonstrated on three large-scale public datasets: NTU-RGB + D, Kinetics, and NTU-RGB + D 120, and also shown in the detailed ablation studies.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1424-8220
1424-8220
DOI:10.3390/s20123499