Weightlifting Performance Analyzer

A Python application that processes weightlifting videos to automatically analyze and provide performance metrics such as rep count, velocity, force applied, and power output.

Features

• Video Upload & Format Support: Upload videos in various formats and resolutions including 4K60fps
• User Input Parameters: Input barbell weight and user height/weight for accurate performance analysis
• Pose Estimation & Motion Tracking: Detect key points on the user's body using MediaPipe
• Automatic Rep Counting: Accurately count repetitions performed during exercises
• Performance Metrics Calculation:
  • Velocity: Measure movement speed during exercises
  • Force Applied: Estimate force exerted based on movement and weights
  • Power Output (Watts): Compute power output combining velocity and force
• Data Visualization: Present analysis results in user-friendly graphs and charts
• Export Results: Save analysis data in CSV format and graphs as images

Supported Exercises

• Squats
• Deadlifts
• Bench Press

Installation

Prerequisites

• Python 3.8 or higher
• pip
• Tkinter (bundled with most Python installs; on Debian/Ubuntu: sudo apt install python3-tk)

Setup

1. Clone this repository:

   git clone https://github.com/arogan178/WeightLifting_Analyser.git
   cd WeightLifting_Analyser

2. Install required dependencies:

   pip install -r requirements.txt

Usage

Graphical User Interface (GUI)

1. Launch the application:

   python main.py

2. Use the GUI to:

   • Open a weightlifting video
   • Set exercise parameters (type, barbell weight, user measurements)
   • Run the analysis
   • View performance metrics and visualizations
   • Save results

Command Line

A headless mode (python main.py --no-gui) is planned but not implemented yet.

Project Structure

├── main.py                  # Main application entry point
├── requirements.txt         # Project dependencies
└── src/                     # Source code
    ├── video/               # Video processing
    │   └── processor.py     # Video frame extraction and preprocessing
    ├── pose/                # Pose estimation
    │   └── estimator.py     # Body keypoint detection and tracking
    ├── metrics/             # Performance metrics
    │   └── calculator.py    # Calculate rep count, velocity, force, power
    ├── visualization/       # Data visualization
    │   └── plotter.py       # Create graphs and charts
    └── ui/                  # User interface
        └── app.py           # GUI implementation

Technical Details

• Computer Vision: Uses OpenCV and MediaPipe for pose detection and tracking
• Physics Calculations: Applies biomechanical principles to calculate performance metrics
• Data Processing: Uses NumPy, Pandas for data manipulation and analysis
• Visualization: Matplotlib and Seaborn for creating visual representations

Requirements

See requirements.txt for detailed dependencies.

License

MIT