Facial-Landmark-Detection

The Objective of this project is to find the best CNN Model capable of predicting 68 Keypoints around the face, eyes, mouth, nose... using Pytorch

Image

  • model.py : contains the model structure.
  • data_load.py : contains classes to create dataset and apply transformations like normalization,cropping and transform the images to tensors.
  • Train.ipynb : This notebook contains the data preprocessing and the training process as well as the test.
  • Face_detection.ipynb : Apply Facedetection using HaarCascade classifiers and detect keypoints in different faces detected in the images.