Movement Estimation Using Mediapipe BlazePose

The paper describes a system to track the body movement of a person from a video source while augmenting the labelled skeleton joints onto the body of the person. This work has endless applications in the real world especially in the physical-demanding working environment as well as in the sports industry by implementing deep learning, the techniques can recognize the joints on a person’s body. An algorithm namely Mediapipe Blazepose has been applied using PoseNet dataset to detect and estimate curated movements specifically designed for body injury during heavy workload. The propose method has been compared to IMU based motion capture and the difference accuracy is within 10% since IMU capture real data of the sensors while the deep learning method using 2D image analysis. The expected outcome from this project is a working system that is able to correctly identify and label the skeleton joints on a person’s body as well as perform various calculation such as movement velocity and the angle of joints which could be crucial for determining whether certain body movements could result in injuries either in the short- or long-term period.