/AI-Body-Language-Decoder

AI Body Language Decoder using MediaPipe

Primary LanguageJupyter Notebook

AI Body Language Decoder

Introduction

The AI Body Language Decoder is a cutting-edge application that leverages the power of MediaPipe to detect and interpret body language in real-time using a webcam. This project utilizes advanced machine learning techniques to analyze pose, facial expressions, and hand gestures, providing insights into the body language of individuals.

Features

  • Real-time detection of body language.
  • Integration with MediaPipe for pose, face, and hand landmark detection.
  • Predictive modeling to classify different body language signals.
  • Visual display of detection results and predictions.

Prerequisites

  • Python 3.x
  • OpenCV
  • MediaPipe
  • Pandas
  • Numpy
  • Pickle

Installation

To set up the AI Body Language Decoder, follow these steps:

  1. Clone the Repository

  2. Install Dependencies

pip install opencv-python

pip install mediapipe

pip install pandas

pip install numpy

  1. Load the Pre-trained Model Ensure you have the body_language.pkl model file in the project directory.

Usage

To run the AI Body Language Decoder, execute the following command:

python run.py

  • The application will activate the webcam.
  • Start performing gestures or poses in front of the camera.
  • The application will display the detected body language class and its probability.

Sample Outputs

  • Face, hand, and pose detection samples:

    Sample Image 1

Acknowledgements

This project is powered by MediaPipe, a robust framework for building multimodal applied machine learning pipelines.