Intelligent fitness system design based on ESP32 and human posture recognition.

Aiming at the fixed and single function of fitness device, this system designs an intelligent fitness system that monitors the human body status in real time by the lower computer and judges whether the fitness action is standard by the upper computer. The system uses ESP32 as the main control chip and obtains environmental information and user's body data in real time through temperature and humidity sensor, pH strength sensor and ECG detection sensor. The lower computer uploads the data to the upper computer through UDP communication protocol and visualizes the data. The upper computer connects to the monocular camera and identifies and determines whether the user's fitness movements are standard or not by using the human posture estimation algorithm in deep learning and gives tips. The user performs the exercise according to the prompts. The system ensures the efficiency of fitness and increases the fun of human-computer interaction.