Multimodal spatiotemporal skeletal kinematic gait feature fusion for vision-based fall detection

作者：

Highlights：

• Presents MSTSK-GFF classifier for detecting falls using video data.

• This framework adopts a Multimodal Feature Fusion (MFF) technique.

• MFF generates spatiotemporal kinematic gait features using STGCN and 1DCNN.

• Spotted hyena optimizer is used to optimize the weights and biases of the network.

• Experimental results have proven the efficiency of the proposed framework.

摘要

•Presents MSTSK-GFF classifier for detecting falls using video data.•This framework adopts a Multimodal Feature Fusion (MFF) technique.•MFF generates spatiotemporal kinematic gait features using STGCN and 1DCNN.•Spotted hyena optimizer is used to optimize the weights and biases of the network.•Experimental results have proven the efficiency of the proposed framework.

论文关键词：Fall risk,Gait patterns,Multimodal feature fusion,Spatiotemporal features,Spotted hyena optimizer,STGCN and 1D-CNN

论文评审过程：Received 7 May 2022, Revised 9 August 2022, Accepted 22 August 2022, Available online 28 August 2022, Version of Record 15 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118681