Human activity recognition using wearable sensors by heterogeneous convolutional neural networks

作者：

Highlights：

• The proposed approach can strengthen the basic convolution.

• Our proposed method divides the filters into two groups but unevenly.

• It can achieve SOTA performance without increasing computational burden.

• The visualizing analysis of heterogeneous convolution is provided.

• Actual operation is evaluated on a Raspberry Pi platform.

摘要

•The proposed approach can strengthen the basic convolution.•Our proposed method divides the filters into two groups but unevenly.•It can achieve SOTA performance without increasing computational burden.•The visualizing analysis of heterogeneous convolution is provided.•Actual operation is evaluated on a Raspberry Pi platform.

论文关键词：Sensor,Grouped convolution,Activity recognition,Deep learning,Heterogeneous convolution

论文评审过程：Received 20 July 2021, Revised 17 January 2022, Accepted 25 February 2022, Available online 11 March 2022, Version of Record 15 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116764