Skip to content

Facial Landmark Detection in Python, using OpenCV for real-time video capture and basic face detection. This project lays the groundwork for an advanced facial landmark model, utilizing a pre-trained face detection model and a custom-trained CNN for accurate facial landmark identification.

License

Notifications You must be signed in to change notification settings

HomaTaherpour/Face-Detection-And-Landmarking

Repository files navigation

Face Detection And Landmarking

Face-Detection-And-Landmarking Face-Detection-And-Landmarking

Introduction

This repository contains scripts tailored for facial landmark detection, offering features like real-time testing, a graphical user interface for landmark selection, image augmentation, and pixel coordinate extraction. The code efficiently employs MobileNetV2 for facial landmark detection, ensuring a balance between model size and accuracy. The model architecture includes a Global Average Pooling layer, dropout, and a Dense layer with sigmoid activation. Noteworthy scripts such as agumantation.py and dimentationtaker.py enable customization based on your images and dataset. Utilize dimentationtaker.py to overlay radio buttons on images and agumantation.py to augment pictures for dataset creation.

Prerequisites

  • Opencv
  • TensorFlow
  • Numpy
  • Imgaug
  • tqdm
  • PIL (Python Imaging Library)
  • Tkinter (for GUI)

dataset

For training and evaluating the model, consider utilizing the lapa dataset, accessible at: [https://github.com/jd-opensource/lapa-dataset]

License

This project is licensed under the MIT License.

About

Facial Landmark Detection in Python, using OpenCV for real-time video capture and basic face detection. This project lays the groundwork for an advanced facial landmark model, utilizing a pre-trained face detection model and a custom-trained CNN for accurate facial landmark identification.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages