Problem Statement

The lack of real-time recognition and response to students' emotional states in classrooms often hinders personalized learning, engagement, and well-being. Therefore, we need an empathetic, privacy-respecting solution that can analyze students' emotional cues for a more supportive learning environment. This study focuses on bridging the gap between educators' understanding of their learners and their social well-being in the classroom environment.

Emotions:

Angry 😠
Disgust 😧
Fear 😨
Happy 😃
Sad 😞
Surprise 😮
Neutral 😐

Dataset link - (https://www.kaggle.com/datasets/ashishpatel26/facial-expression-recognitionferchallenge)

What We Offer

Real-time detection and interpretation of students' facial expressions.
Valuable insights into emotional engagement, comprehension levels, and overall well-being.
Enabled and tailored teaching strategies fostering an empathetic and inclusive learning environment.
Comprehensive understanding of classroom dynamics and educational insights.
Continuous updates and timely feedback to learners.

A Fresh Approach

Instead of solely focusing on recognizing facial emotions, we take facial emotion recognition to the next level by creating an interactive emotion-based learning environment. Our goal is to build a system that can detect students' emotions and dynamically adapt classroom dynamics to create a more engaging and supportive learning experience, thereby transforming the landscape of educational platforms.

Features

Emotion Detection
Emotion Identification
Personalization - Student-Teacher Approach
Engagement Tracking
Mental Health Monitoring and Inclusivity
Real-time Feedback
Data-Driven Insights
Collaborative Learning
Room for Action Research

Methodology / Approach

These are the steps involved in making this project:

Importing Libraries
Data Importing
Data Exploration
Data Configuration
Preparing the Data
Creating a Generator for Training Set
Creating a Generator for Testing Set
Writing the labels into a text file 'Labels.txt
Model Creation
Model Compilation
Training the Model (batch_size = 32, epochs = 100)
Testing Predictions
Saving model as 'model.h5'
Deploying the Model as a Web Application using Streamlit

List of API Toolkits/Libraries

oneDNN
oneMKL
oneDAL
OpenVINO

Technologies Used

Pandas
Numpy
Tensorflow
OpenCV
Keras
TensorFlow Hub

Repository

https://github.com/nemoNehaM/mindmentors 

Conclusion

The current project is the prototype of the more significant project that can be applied in the real-time classroom. The project aims to implement a classroom web camera to monitor the learners' social behavior and learning patterns.
The project also aims at providing with zooming facility the understand any individual.
The real-time web application would work by slightly modifying the current prototype and aiming to capture the entire classroom. 
From the web applications end, the data can also be analyzed using a web camera.

For further details and theoretical explanation, kindly read through the GitHub Repository.