ARC-Net: Activity Recognition Through Capsules

TL;DR

ARC-Net introduces a capsule-based deep learning model for human activity recognition that effectively fuses multiple IMU data sources, outperforming previous methods with a 2% accuracy increase and providing insights into data utilization.

Contribution

The paper presents ARC-Net, a novel capsule network architecture that improves HAR accuracy by better integrating multi-IMU data and visualizing data source contributions.

Findings

Achieved 2% higher accuracy than state-of-the-art methods.

Visualized data source utilization via heatmaps.

Analyzed activity confusion matrices for better understanding.

Abstract

Human Activity Recognition (HAR) is a challenging problem that needs advanced solutions than using handcrafted features to achieve a desirable performance. Deep learning has been proposed as a solution to obtain more accurate HAR systems being robust against noise. In this paper, we introduce ARC-Net and propose the utilization of capsules to fuse the information from multiple inertial measurement units (IMUs) to predict the activity performed by the subject. We hypothesize that this network will be able to tune out the unnecessary information and will be able to make more accurate decisions through the iterative mechanism embedded in capsule networks. We provide heatmaps of the priors, learned by the network, to visualize the utilization of each of the data sources by the trained network. By using the proposed network, we were able to increase the accuracy of the state-of-the-art approaches by 2%. Furthermore, we investigate the directionality of the confusion matrices of our results and discuss the specificity of the activities based on the provided data.