Computer Vision-Driven Evaluation System for Assisted Decision-Making in Sports Training

Computer vision has become a fast-developing technology in the field of artificial intelligence, and its application fields are also expanding, thanks to rapid development deep learning. It will be great practical value if it is combined with sports. When traditional exercise assistance system introduced into sports training, athlete’s training information can obtained by monitoring process through sensors other equipment, which assist athlete retrospectively analyzing technical actions. However, must equipped multiple sensor devices, provided accurate. This paper proposes motion evaluation based on learning algorithms for human posture recognition. The divided three sections: standard database, auxiliary instruction, overall evaluation. database customized user, teaching integrated. user’s actions compared intuitively displayed trainers as data. system’s component recognize display video files, giving an intelligent platform. Simulator tests available. demonstrates efficacy algorithm used this paper.