TY  - INPR
KW  - Convolution

KW  - 
Adaptation models

KW  - 
Topology

KW  - 
Deformable models

KW  - 
Correlation

KW  - 
Convolutional neural networks

KW  - 
Laplace equations
A1  - Myung, Woomin
A1  - Su, Nan
A1  - Xue, Jing-Hao
A1  - Wang, Guijin
PB  - Institute of Electrical and Electronics Engineers (IEEE)
JF  - IEEE Transactions on Image Processing
N1  - This version is the author accepted manuscript. For information on re-use, please refer to the publisher?s terms and conditions.
UR  - http://dx.doi.org/10.1109/tip.2024.3378886
ID  - discovery10189813
N2  - Graph convolutional networks (GCN) have recently been studied to exploit the graph topology of the human body for skeleton-based action recognition. However, most of these methods unfortunately aggregate messages via an inflexible pattern for various action samples, lacking the awareness of intra-class variety and the suitableness for skeleton sequences, which often contain redundant or even detrimental connections. In this paper, we propose a novel Deformable Graph Convolutional Network (DeGCN) to adaptively capture the most informative joints. The proposed DeGCN learns the deformable sampling locations on both spatial and temporal graphs, enabling the model to perceive discriminative receptive fields. Notably, considering human action is inherently continuous, the corresponding temporal features are defined in a continuous latent space. Furthermore, we design an innovative multi-branch framework, which not only strikes a better trade-off between accuracy and model size, but also elevates the effect of ensemble between the joint and bone modalities remarkably. Extensive experiments show that our proposed method achieves state-of-the-art performances on three widely used datasets, NTU RGB+D, NTU RGB+D 120, and NW-UCLA.
AV  - public
TI  - DeGCN: Deformable Graph Convolutional Networks for Skeleton-Based Action Recognition
Y1  - 2024///
ER  -