😃 Emotional Facial Recognition

📌 Project Overview

This project focuses on emotion recognition using deep learning, leveraging ResNet50 as the base model. It classifies facial expressions into 8 emotion categories, including:

• 😊 Happiness
• 😢 Sadness
• 😡 Anger
• 😲 Surprise
• 😨 Fear
• 😏 Contempt
• 🤢 Disgust
• 😐 Neutral

🚀 Features

✅ Dataset Preprocessing – Data cleaning, augmentation, and feature extraction ✅ Deep Learning Model – Transfer Learning using ResNet50 ✅ Classification Performance – Achieves high accuracy using categorical cross-entropy loss ✅ Model Optimization – Includes Early Stopping, Learning Rate Scheduling, and Dropout Regularization ✅ Evaluation Metrics – Confusion Matrix, Precision, Recall, and F1-Score ✅ Data Visualization – Age distribution, gender ratio, and emotion frequency analysis ✅ Deployment Ready – Model saved using TensorFlow's save_model()

🏗️ Tech Stack

Libraries & Frameworks

• TensorFlow / Keras
• OpenCV
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-learn

Machine Learning Techniques

• Transfer Learning with ResNet50
• Image Augmentation using ImageDataGenerator
• Feature Scaling and Label Encoding

📊 Model Performance

• Training Accuracy: 99%
• Validation Accuracy: 95%
• Test Accuracy: 96%
• Loss: Low categorical cross-entropy loss

🙌 Acknowledgments

• Inspired by advancements in computer vision & deep learning
• Built using TensorFlow, OpenCV, and ResNet50

💡 Feel free to contribute, provide feedback, or use this for your own research! 🚀