Hierarchical Multi-Scale Adaptive Conv-LSTM Network for Human Action Recognition Based on Wearable Sensors

This is the authors' abstract. We don't add key points for this paper.

Recently, human action recognition has been widely used in the fields of health monitoring, human-robot interaction, medical treatment, and sports. Due to the availability of various wearable devices on the market, we can easily access sensor data for human action recognition. However, it is still a challenge to capture minute action processes as well as extract spatio-temporal motion patterns from serial sensor data. Therefore, we propose a novel hierarchical multi-scale adaptive Conv-LSTM network structure called HMA Conv-LSTM. The finer-grained spatial information in the sensor signals is extracted by hierarchical multi-scale convolution. The multi-channel feature fusion through adaptive channel feature fusion retains important information and improves model efficiency. We capture temporal context information by dynamic channel selection-LSTM based on the attention mechanism. Extensive experiments on the Opportunity and PAMAP2 public datasets show that our proposed model achieves competitive performance compared to several state-of-the-art approaches.