This Machine Learning Web Application utilizes a Convolutional Neural Network to process the Images, detect and predict the facial Expression. The Dataset to process the Deep Learning Algorithm is taken from FER 2013 dataset from Kaggle. This trained model has 63% accuracy in Facial Expression Recognition. This application uses Haar Cascade Classifier for detecting the faces.
- Drop a ⭐ on the Github Repository.
- Clone the Repo by going to your local Git Client and pushing in the command:
https://github.com/121yaseen/FacialExpressionIdentifier.git
-
Install the required Packages:
-
At last, push in the command:
python main.py
- Go to
http://127.0.0.1:5000/
and enjoy the application.
- It uses webcam for video input default. You can predict on custom videos by changing the video source in camera.py
- Samples videos for prediction are available Here
- Model has 63% accuracy which is not a state-of-the-art performance
- Set up an HTML form to upload custom videos for prediction
- Deploying the Web Application on Cloud.
- Development of an architecture using Pre-Trained Model like VGG16.
- Implementing the Model in PyTorch.
- Enhance the User-Interface using HTML/CSS.
- Set the Application on Docker.