A Stereo Vision Method for IMU-based Motion Tracking Systems Calibration

This paper proposes a new calibration method, based on stereo vision, for the position alignment of Inertial Measurement Unit (IMU) sensors, used in human motion tracking systems. Compared to other methods, the proposed one does not rely on a movement, and therefore, it does not depend on the capability of the person of performing the movement properly. The method is based on the acquisition of a static stereo image of the person to be tracked, while he/she is wearing the IMU sensors and visual markers are placed on both the IMU sensors and some landmark points on the body joints. From the stereo image, the lengths of the body segments and the rotation matrices to align the sensor orientations to the body segments are evaluated. These values are used by the motion tracking algorithm to get the actual segment orientations from the IMU data. The presented calibration method is validated by comparing the obtained measurements with an inclinometer and a laser distance meter.