This repository is made up of files form the Udacity Computer Vision, Nanodegree. This project explores fundamental knowledge of Deep Learning Architectures and Image processing techniques to construct a model that detects facial keypoints (68 in total) around the eyes, nose, mouth and the over all shape of the face. This project is fundamental to emotion recognition, facial tracking, facial pose recognition, and facial filters.
- Detect all faces using Haar Cascade Classifiers using OpenCV
- Detect facial keypoint with a Convolutional Neural Network
conda create -n cv-nd python=3.6
source activate cv-nd
conda install pytorch torchvision cuda91 -c pytorch
pip install -r requirements.txt
wget -P ./data/ https://s3.amazonaws.com/video.udacity-data.com/topher/2018/May/5aea1b91_train-test-data/train-test-data.zip
cd data
unzip train-test-data.zip
rm train-test-data.zip
The project will be broken up into a few main parts in four Python notebooks:
Notebook 1 : Loading and Visualizing the Facial Keypoint Data
Notebook 2 : Defining and Training a Convolutional Neural Network (CNN) to Predict Facial Keypoints
Notebook 3 : Facial Keypoint Detection Using Haar Cascades and your Trained CNN
Notebook 4 : Fun Filters and Keypoint Uses