The invention relates to the technical field of sports facilities, in particular to an AI
standing long jump evaluation system supporting automatic calibration. Automatic calibration is achieved through the Arcomarkers check code square pad and the high-definition camera,
long jump pad laying and scale drawing are not needed, convenience and high efficiency are achieved, the site construction and maintenance cost is reduced, the method is not limited by site conditions, the method can be used in indoor and outdoor irregular sites, the
system universality is greatly improved, and in the process from the take-off moment to the landing moment, the
system reliability is improved. The relative distance between a take-off point and a
landing point can be accurately calculated by detecting
heel key points in real time, analyzing the motion trail of
lower limb key points and accurately determining a take-off frame and a landing frame in combination with a virtual coordinate
system and scale marks, and therefore the accurate
standing long jump score is obtained. According to the method, measurement errors are effectively reduced, the accuracy and fairness of
score evaluation are improved, and the strict requirements of professional sports events, athlete training, school
physical education and the like for
score precision can be met.