The use of artificial intelligence and computer vision technologies for automating fitness exercise monitoring using the example of the online platform FORA VISION

The relevance of automating exercise monitoring is due to the increasing need for online fitness programs, the use of artificial intelligence to analyze exercise techniques and encourage users to engage in fitness classes through convenient digital platforms that contributing to the effectiveness of training. Objective. Development and implementation of a technical solution for monitoring physical activity during training sessions using a mobile phone and a computer using artificial intelligence and computer vision technologies. Methods. The proposed solution used the MediaPipe Pose Detection computer vision framework, whose neural networks are pre-trained on a large amount of data and can accurately handle the task in situations with different backgrounds, clothes and people. Results. A technical solution has been developed - the FORA VISION online platform (https://fora.vision), which is based on neural networks trained on a labeled dataset of physical exercises, which allows you to automatically track the performance of exercises and calculate their number. Conclusion. Motion recognition and evaluation technology is applicable in various fields, from conducting distance physical education classes in educational institutions to therapeutic and industrial gymnastics, as well as for self-development of basic movement skills in various sports and professionally applied fields.