Automatic Synchronization of Markerless Video and Wearable Sensors for Walking Assessment

As walking assessment is commonly done through visual inspection, it is beneficial to make joint angle information available to the clinicians for better assessment. The main challenge is that the video camera and inertial sensor are usually two separate systems, and the recordings are hard to be initialized at the same time manually. This creates a problem that the inertial sensor data is not temporally synchronized with the video camera. This paper proposes a method to synchronize the video and sensor data by detecting and matching the maximum backward swings of the leg. The proposed method is validated by blinking LED and transmitting LED flag to the computer at the same time. The synchronization error of the proposed method is low at about 0 ± 2 frames compared to the validation method.