FitNex - AI Powered Fitness Tracker

A project developed as part of the AICTE Internship

---

About the Project

FitNex is an AI-powered workout assistant designed to help users track their workout repetitions accurately. Using computer vision techniques and real-time pose estimation, the app analyzes workout videos or live webcam feeds to provide feedback during exercise sessions. This project is the foundation for a full-fledged AI fitness trainer application, with plans to add advanced features and an improved user interface.

---

Features

• Automatically detects arm movements and counts bicep curl repetitions.
• Works with both pre-recorded videos and live webcam streams.
• Displays the count of left and right arm bicep curls live on the screen.
• Visualizes key landmarks (shoulder, elbow, wrist) and angles during the workout.
• Has user-friendly Streamlit interface for uploading videos or analyzing workouts via webcam.
• Powered by the YOLOv8 pose model for precise keypoint detection.

---

Setup Instructions

1. Clone the Repository

Use the following commands to clone the repository and navigate to the project directory:

git clone https://github.com/<your-username>/AI_Fitness_Tracker-AICTE_Internship.git
cd AI_Fitness_Tracker-AICTE_Internship

2. Install Dependencies

Ensure you have Python installed (version 3.8+ recommended). Then, install the required libraries:

pip install -r requirements.txt

3. Run the Application

Launch the Streamlit app using the following command:

 streamlit run app.py

How to Use

1. Select Input:

   • Choose between Upload Video or Use Webcam in the application interface.

2. For Upload Video:

   • Click on Upload Video and select a workout video (e.g., bicep curl video) for analysis.

3. For Webcam:

   • Click on the Start Webcam button to start real-time tracking.

4. View Results:

   • The app will display live feedback, including:
     • Left and right bicep curl counts.
     • Visual landmarks (shoulder, elbow, wrist) and connecting lines.

---

Technology Stack

• Python: Core programming language for backend logic.
• YOLOv8: Pose estimation model for detecting keypoints with precision.
• OpenCV: Real-time video processing and visualization.
• Streamlit: Interactive web-based interface for seamless user interaction.
• NumPy: Mathematical computations for angle calculations.

---

Contact

For questions or feedback, feel free to reach out via:

• Email: deekshitha1325@gmail.com
• GitHub: https://github.com/deekshitha-3

This project serves as the foundation for a comprehensive AI Fitness Trainer Project - FitNex.