TY - GEN
T1 - Variant Graph Convolutional Networks for Skeleton-Based Hand Action Recognition
AU - Htwe, Khin Sabai
AU - Watanabe, Hiroshi
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Graph convolutional network is widely used in skeleton-based applications such as action recognition. In this paper, a Variant Graph Convolutional Network (VGCN) is proposed to learn not to be constrained of the physical connections of hand structure since a predefined fixed graph structure lacks of flexibility to capture variance and different actions. With experiment on our hand actions skeleton dataset, the proposed method outperform with significance accuracy to the conventional ones.
AB - Graph convolutional network is widely used in skeleton-based applications such as action recognition. In this paper, a Variant Graph Convolutional Network (VGCN) is proposed to learn not to be constrained of the physical connections of hand structure since a predefined fixed graph structure lacks of flexibility to capture variance and different actions. With experiment on our hand actions skeleton dataset, the proposed method outperform with significance accuracy to the conventional ones.
KW - graph convolutional network
KW - hand action
KW - recognition
KW - skeleton information
KW - variant joints
UR - http://www.scopus.com/inward/record.url?scp=85123478171&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123478171&partnerID=8YFLogxK
U2 - 10.1109/GCCE53005.2021.9621997
DO - 10.1109/GCCE53005.2021.9621997
M3 - Conference contribution
AN - SCOPUS:85123478171
T3 - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021
SP - 651
EP - 652
BT - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th IEEE Global Conference on Consumer Electronics, GCCE 2021
Y2 - 12 October 2021 through 15 October 2021
ER -