The research aim was to analyse the influence of velocity and size of markers on the accuracy of motion capture measurement utilising image processing with the use of OpenCV. On the basis of the obtained results, the usefulness of the applied measurement method in studying the kinematics of the human body while driving operating a wheelchair was determined. This article presents the test results for a low-budget motion capture measurement system for testing the kinematics of the human body in a single plane. The tested measuring system includes a standard activity camera Xiaomi Yi4K, expanded polystyrene markers with printed ArUco codes, and original software for marker position detection developed by the author. The analysis of the measurement method with regard to its applicability in biomechanical studies has highlighted several key factors: the number of measuring points, measurement accuracy expressed as a relative error and the limit velocity at which the marker trajectory is correctly represented. The article shows that the limit velocity of the marker is 2.2 m/s for 50x50 mm markers and 1.4 m/s for 30x30 mm markers. The number of measured points ranged from 233 to 2,457 depending on the marker velocity. The relative error did not exceed 5% for the marker velocities and thus provided a correct representation of its trajectory.