Home
Softono
Smart-AI-Gym-Trainer-With-RealTime-Exercise-Tracking-and-Nutrition-Assistant

Smart-AI-Gym-Trainer-With-RealTime-Exercise-Tracking-and-Nutrition-Assistant

Open source PHP
GitHub
11
Stars
1
Forks
0
Issues
0
Watchers
9 months
Last Commit
Workout & Activity Tracking AI & Machine Learning Nutrition & Diet

About Smart-AI-Gym-Trainer-With-RealTime-Exercise-Tracking-and-Nutrition-Assistant

AI Gym Trainer is an all-in-one smart fitness platform that combines computer vision and artificial intelligence to deliver real-time exercise tracking, posture correction, and dietary analysis. Using OpenCV, MediaPipe, and a custom pose estimation model, the system detects seven exercises including squats, push-ups, and dumbbell curls, providing instant feedback on form with messages such as Good Posture or Fix Your Form. A built-in YOLOv5 based food recognition module estimates nutritional content from camera-captured images, enabling smart diet tracking. The application calculates key health metrics like BMI and BMR, stores user data in a MySQL database, and generates personalized workout routines and meal plans tailored to individual body profiles and goals such as weight loss, muscle gain, or maintenance. The platform features a web interface built with HTML, CSS, and PHP alongside a Python backend using Flask or Streamlit and a Tkinter desktop option. It also includes badges, achievement tracking, macro

Platforms

Web Self-hosted

Languages

PHP

Links

Source Code
p
Published by
preethamn15
Visit View Profile

README.md

View on GitHub

🤖 AI Gym Trainer

Your Smart Workout & Nutrition Assistant


Real-Time Exercise Tracking · Posture Feedback · AI Nutrition Planner

Overview

AI Gym Trainer is an all-in-one smart fitness platform designed to revolutionize home and personal workouts using the power of Artificial Intelligence and Computer Vision. This intelligent system offers real-time monitoring of exercises, automated posture correction feedback, and advanced dietary analysis—making it a comprehensive virtual gym assistant.

Whether you're a beginner looking for guidance, a student working on a final year project, or a fitness enthusiast aiming to optimize your training, this platform is built to help you achieve your goals with minimal equipment and maximum efficiency.

Features

  • Live Exercise Tracking – Detects 7 exercises using computer vision (e.g., squats, push-ups, dumbbell curls, alternative dumbbell curls).
  • Posture Feedback – Real-time form evaluation: Good Posture or Fix Your Form.
  • Food Recognition – YOLOv5 model estimates nutrition from camera-captured food images.
  • Health Metrics – Calculates BMI, BMR, and stores data in a MySQL database.
  • Personalized Fitness Plan – Tailored routines based on your body profile and goals.
  • Meal Planner – Smart meal suggestions for weight loss, muscle gain, or maintenance.
  • AI Integration – Uses deep learning & pose estimation for accurate analysis.
  • MySQL Integration – Logs workouts, meals, and progress for every user.

Tech Stack

Area Technologies
Computer Vision OpenCV, MediaPipe, YOLOv5
AI/ML Custom Pose Estimation, Torch Models
Frontend/UI HTML, CSS, PHP (Web), Tkinter
Backend Python (Flask or Streamlit - customizable)
Database MySQL
IDE VS Code

Use Cases

  • Home workouts with real-time feedback
  • Smart diet tracking using food detection
  • Fitness planning for students, professionals & beginners
  • Great Final Year Project with AI + ML integration

Sample Screens & Demo

🎥 Demo Video – Full Walkthrough

Watch Demo

đź”— Click the image above download the .mov demo video you will get the full view of website.

AI Personalized Plan

Badges and Achievements

Body Metrics Calculator

🍽Macronutrient Calculator

This is our final year project, and it includes all the trending technologies like AI and Deep Learning. For 1st, 2nd, and 3rd-year students, this can be a great reference model to showcase in your college and score full marks.

đź“„ Research Paper

We have also published a research paper based on an earlier version of this project — feel free to check it out for reference.
Since then, we've made significant improvements to the model.

View Research Paper (PDF)

⚠️ Note: I won’t be uploading the main folder and SQL file required to run the full model here. 📩 If you’re interested in running the complete project, feel free to contact me — I’ll provide all the necessary files and step-by-step guidance.

Contact 📧 [email protected] 📧 [email protected]