Introduction. The article presents the areas of use of computer vision technologies in the field of physical education and sports activities. Methodology. Innovative methods for assessing motor action based on video material, pattern recognition, object tracking are analyzed, the possibility of automated control of motor action based on the use of artificial intelligence and neural networks is shown. Particular attention is paid to the possibilities to integrate computer vision technologies from the field of professional sports and fitness industry into fitness physical culture, educational process. Results. A computer program based on the MediaPipe system has been developed for automated assessment of the effectiveness of physical exercise based on biomechanical properties of motor action, taking into account individual proportions of the body, determining the load on key points of the body in order to obtain an optimal health effect. Conclusions. The prospects for using this program to optimize the methodology of health-improving physical education are discussed.