SpiderNet: A spiderweb graph neural network for multi-view gait recognition

作者：

Highlights：

• A spiderweb graph neural network is built for multi-view gait recognition.

• The SpiderNet learns the spatio-temporal and structural knowledge of the gait.

• The SpiderNet can integrate the high and low frequency features in gait data.

• The SpiderNet fuses features of various views and outperforms other methods.

摘要

•A spiderweb graph neural network is built for multi-view gait recognition.•The SpiderNet learns the spatio-temporal and structural knowledge of the gait.•The SpiderNet can integrate the high and low frequency features in gait data.•The SpiderNet fuses features of various views and outperforms other methods.

论文关键词：Multi-view,Gait knowledge,Spiderweb graph convolutional network,Feature fusion,Capsule network,Long-short term memory,Spatio-temporal information

论文评审过程：Received 9 March 2020, Revised 13 July 2020, Accepted 15 July 2020, Available online 27 July 2020, Version of Record 7 August 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106273