---
title: Facial Emotion Recognition
emoji: 😊
colorFrom: blue
colorTo: pink
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false
---

# 😊 Facial Emotion Recognition (EfficientNet-B4 + Focal Loss + MixUp/CutMix)

This project demonstrates **Facial Emotion Recognition** using a fine-tuned **EfficientNet-B4** model trained on a balanced subset of the **AffectNet** dataset.

It can detect the emotion of a person from:
- 📸 **Uploaded images**
- 🎥 **Live camera feed**

---

## 🧠 Model Overview

| Feature | Description |
|----------|--------------|
| **Base Model** | EfficientNet-B4 (from `timm` library) |
| **Loss Function** | Focal Loss (γ=2, α=1) |
| **Regularization** | MixUp & CutMix augmentations |
| **Optimizer** | AdamW with CosineAnnealingLR |
| **Input Size** | 224×224 |
| **Classes** | Anger, Contempt, Disgust, Fear, Happy, Neutral, Sad, Surprise |

These techniques were introduced to reduce **overfitting** and enhance model **generalization**, achieving higher validation accuracy compared to a baseline model using standard CrossEntropyLoss.

---

## 📊 Performance Summary

| Metric | Baseline (CE Loss) | Improved Model (Focal + MixUp/CutMix) |
|--------|--------------------|--------------------------------------|
| Train Accuracy | 97.8% | 60.6% |
| Val Accuracy | 77.2% | **79.2%** |
| Val F1 Score | 0.77 | **0.79** |
| Early Stopping | No | ✅ Yes |
| Overfitting | ❌ Present | ✅ Reduced |

> Note: The validation accuracy surpasses the training accuracy in the improved model, indicating successful **regularization** and **harder training augmentations** that improve generalization.

---

## ⚙️ Usage

### 📸 Upload Image
Upload an image containing a face — the model will detect the face, draw a bounding box, and display the predicted emotion label.

### 🎥 Live Camera
Activate your webcam to detect emotions in real-time. The bounding box and emotion label are updated continuously.

---

## 🧩 Tech Stack

- **Python 3.10+**
- **PyTorch** (for model training & inference)
- **Timm** (for EfficientNet backbone)
- **OpenCV** (for face detection)
- **Gradio** (for the web interface)
- **NumPy**, **Pillow**, **Seaborn**, **Matplotlib** (for visualization & preprocessing)

---

## 📂 Dataset

The model was trained using a **balanced subset of the AffectNet dataset**, resized to 224×224 with approximately 3,690 images per class.  
Augmentations included random flips, color jitter, and light rotations to preserve facial expression semantics.

---

## 👨‍💻 Authors

Developed by:
- **Abdelrhman Khaled Hassan**

As part of an academic research project exploring **emotion classification and regularization techniques** in deep learning.

---

## 📚 References

- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946)
- [Focal Loss for Dense Object Detection (Lin et al., 2017)](https://arxiv.org/abs/1708.02002)
- [MixUp: Beyond Empirical Risk Minimization](https://arxiv.org/abs/1710.09412)
- [CutMix: Regularization Strategy to Train Strong Classifiers](https://arxiv.org/abs/1905.04899)
- [AffectNet: A Database for Facial Expression, Valence, and Arousal Computing in the Wild](https://arxiv.org/abs/1708.03985)

---

### 🌐 Live Demo

Once deployed, your live demo will be available at:
> https://huggingface.co/spaces/abdelrhman145/Face_Emotion_Recognition

---