Human Motion Monitoring Platform Based on Positional Relationship and Inertial Features

Abstract

Human motion monitoring widely used in medical rehabilitation, health surveillance, video effects, virtual reality and natural human–computer interaction. Existing human motion monitoring can be divided into two ways, non-wearable and wearable. Non-wearable human motion monitoring system can be used when it is not contact with user, so it has no influence on users’ daily life, but its monitoring range is constrained. Wearable human motion monitoring system can solve this problem well. But as the dataset is not complete, it is inconvenient for the research of related algorithms, and many researchers choose to build their own data acquisition platform. On the other hand, the traditional motion recognition platform which is based on inertial features, is difficult to identify static pose. This limits the application of wearable motion recognition technology. In this paper, we design a human motion recognition platform based on inertial features and positional relationship, which can provide a platform for data acquisition and dataset for the researchers who study algorithm of wearable motion recognition.