Automated Real-Time Facial Expression Recognition

Project Title

Automated Real-Time Facial Expression Recognition

Project Area of Specialization Artificial IntelligenceProject Summary

This project “Automated Real-Time Facial Expression Recognition” help us to recognize seven universal facial human expressions (Angry, Sad, Happy, Fear, Disgust, Neutral, and Surprise) by a mobile app in real-time using your mobile camera.

Project Objectives

The main objective of this project is to design a model for Real-time Facial Expression Recognition using Machine Learning and Deep Learning and deploying in mobile app.

Project Implementation Method

This project uses dataset from “Challenges in Representation Learning: A report on three machine learning contests” research which consists of 48x48-pixel grayscale images of face. The size of dataset is 35,887. The model will classify and give output in the seven universal facial human expressions (Angry, Sad, Happy, Fear, Disgust, Neutral, and Surprise). The model will develop using python and mobile app will develop using react-native.

We are planning to use following libraries and frameworks,

• Tensor Flow and Keras used for creating models, neural networks design, Image Classification, and etc.
• OpenCV used for Image Processing including Face detection.
• Matplotlib and Seaborn used for Data Visualization.
• Numpy and Pandas used for data cleaning, data exploration and data analysis.
• React-Native for deployment of feature in a mobile app.

Benefits of the Project

The ability to understand facial expressions is really important for better communication. If you only listen what person is saying and ignoring their expressions you can’t judge which mood they are in and can’t communicate effectively with them. Sometimes words are not powerful as of expressions. So we aim to build a mobile app in which it can automatically detect the facial expressions in real-time environment. It is more than handy and can have vast areas of implementation. For example, it can be implemented for blind people or visually impaired people, in restaurants for analyzing customer reviews, and many more.

Technical Details of Final Deliverable

Final output (facial expresstion) of this project will be seen in a mobile app in real-time using your mobile camera.

Final Deliverable of the Project Software SystemCore Industry ITOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Partnerships to achieve the GoalRequired Resources