Gaze Correction

Gaze Correction is a powerful deep learning-based solution for correcting the gaze direction in images and videos. This repository provides a comprehensive implementation of a gaze correction model, along with tools for training, evaluating, and deploying the model.

Table of Contents

• Features
• Installation
  • Prerequisites
  • Virtual Environment Setup
  • Installing Dependencies

Installation

Prerequisites

• Python 3.9 or higher
• pip package manager
• virtualenv (optional, but recommended)

Required Packages

• tensorflow 2.15.0
• tensorflow-probability 0.23.0
• dlib
• opencv-python

Virtual Environment Setup (Recommended)

It's recommended to use a virtual environment to manage the project's dependencies. Follow these steps to set up a virtual environment:

1. Install virtualenv if you haven't already:

   pip install virtualenv

2. Install virtualenv if you haven't already:

   virtualenv .venv

3. Activate the virtual environment:

   • On Windows:

     .venv\Scripts\activate

   • On macOS/Linux:

     source .venv/bin/activate

Installing Dependencies

1. Clone the repository:

   git clone https://github.com/dtoyoda7/Gaze_Correction.git
   cd Gaze_Correction

2. Install the required dependencies:

   pip install -r requirements.txt

   This will install all the necessary Python packages, including TensorFlow 2.x and other libraries required for the gaze correction project.

3. (Optional) Install additional development dependencies:

   pip install -r requirements-dev.txt

   This will install additional tools and libraries for development, such as linters, formatters, and testing frameworks.

Now you're ready to start using the gaze correction project. Proceed to the Usage section to learn how to train, evaluate, and use the gaze correction model.