Back to Search Start Over

Rank-GCN for Robust Action Recognition

Authors :
Haetsal Lee
Unsang Park
Ig-Jae Kim
Junghyun Cho
Source :
IEEE Access, Vol 10, Pp 91739-91749 (2022)
Publication Year :
2022
Publisher :
IEEE, 2022.

Abstract

We present a robust skeleton-based action recognition method with graph convolutional network (GCN) that uses the new adjacency matrix, called Rank-GCN. In Rank-GCN, the biggest change from previous approaches is how the adjacency matrix is generated to accumulate features from neighboring nodes by re-defining “adjacency.” The new adjacency matrix, which we call the rank adjacency matrix, is generated by ranking all the nodes according to metrics including the Euclidean distance from the nodes of interest, whereas the previous GCNs methods used only 1-hop neighboring nodes to construct adjacency. By adopting the rank adjacency matrix, we find not only performance improvements but also robustness against swapping, location shifting and dropping of certain nodes. The fact that the human-made rank adjacency matrix wins against the deep-learning-based matrix, implies that there are still some parts that need touch of humans. We expect our Rank-GCN can make performance improvements especially when the predicted human joints are less accurate and unstable.

Details

Language :
English
ISSN :
21693536
Volume :
10
Database :
Directory of Open Access Journals
Journal :
IEEE Access
Publication Type :
Academic Journal
Accession number :
edsdoj.6831d2c88ec545c98886cbf6d2e7d824
Document Type :
article
Full Text :
https://doi.org/10.1109/ACCESS.2022.3202164