Capturing Motion,
Crafting Stories

With the fast growing of sports science and sports engineering, more attention has been paid to the application of combining sports with science and technology. Motion capture technology can obtain kinematic parameters accurately, its performance has been recognized and used in sports training continuously. Obtaining accurate kinematic parameters is not only beneficial to the statistics of motion laws, but also standardizes physical education and coaching, so as to get rid of the previous method which entirely depends on experience.

College of Physicial Education, TUT has applied motion capture technology to basketball coaching. The researchers used NOKOV optical motion capture system to obtain the 3D coordinates of the markers on the athlete in real time, which was used to analyze the movements of the athlete at various stages from a 3D perspective, and to create the kinematics model of whole body. In order to recognize human posture efficiently and accurately, a posture analysis method based on the similarity matching between feature planes was proposed.

Firstly, NOKOV motion capture system was used to collect the real-time data of human motion feature points, and the feature plane was determined according to the feature points, and then the angle between the feature vector and the attitude feature vector was extracted. Next, according to the movement characteristics of the key parts of the basketball movement, the characteristic correlation coefficient of human posture is calculated. After that, the difference between student basketball movement and standard movement is compared by the correlation coefficient of feature vector and its included angle in order to improve the accuracy. This method can effectively solve the calculation error caused by the inherent characteristics of the object to be measured, and reduce the computational complexity, so as to improve the efficiency and stability of posture analysis.

After capturing a large amount of motion data, the researchers established a basketball motion database for teaching. The 3D data obtained by motion capture is bound to the human body model in the computer to create tutorial animation. By watching the 3D animation, athletes could quickly find their own problems, and correct their movements.

The digitalized physical education integrates NOKOV motion capture to realize real-time interaction between real and virtual scenes. Therefore, the previous physical education teaching mode that relies on experience has been upgraded to a vivid and intuitive intelligentization of education. Teachers could make quantitative analysis of movements more convenient and efficient to help students improving.