Sequential inter-hop graph convolution neural network (SIhGCN) for skeleton-based human action recognition

作者：

Highlights：

• A graph convolution model for skeleton-based action recognition is proposed.

• Normalized Laplacian Matrix is utilized to encode the graph information.

• An attention-based feature aggregation is proposed to extract the salient features.

• The proposed method achieves better results than the baseline models.

摘要

•A graph convolution model for skeleton-based action recognition is proposed.•Normalized Laplacian Matrix is utilized to encode the graph information.•An attention-based feature aggregation is proposed to extract the salient features.•The proposed method achieves better results than the baseline models.

论文关键词：Action recognition,Attention mechanism,Feature aggregation,Graph convolutional neural network,Normalized Laplacian matrix

论文评审过程：Received 30 July 2020, Revised 20 November 2020, Accepted 17 January 2022, Available online 22 January 2022, Version of Record 2 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116566