Facial Emotion Recognition (FER Project)

This project performs real-time facial emotion detection using a CNN model.
It supports both:

• OpenCV (desktop webcam)
• Web browser (FastAPI + frontend)
• Web Browser Version With Shell
• Vercel and Render For Web Deployment (TBD)

---

📁 Project Setup and Structure

FER-PROJ-2/
├── .github/
│   ├── workflows/
│   ├── ci.yml
├── api/
│   ├── app.py
│   ├── __init__.py
├── data/
│   ├── processed/
│   │   ├── train/
│   │   └── validation/
│   └── raw/
│       ├── test/
│       └── train/
├── frontend/
│   └── index.html
│   ├── ai.html
│   ├── realtime.html
├── tests/
│   ├── test_api.py
│   ├── test_data.py
│   ├── test_imports.py
│   ├── test_inference.py
│   ├── test_model.py
├── models/
│   ├── best_emotion_model.keras
│   └── final_emotion_model.keras
└── src/
    ├── data/
    │   ├── __pycache__/
    │   ├── __init__.py
    │   ├── preprocess.py
    ├── inference/
    │   ├── __init__.py
    │   ├── predict.py
    │   └── webcam.py # if user prefers openCV
    ├── llm/
    │   ├── __init__.py
    │   ├── groq_client.py
    ├── models/
    │   ├── __init__.py
    │   ├── evaluate.py  # MLFLOW included
    │   ├── model.py
    │   └── train.py     # MLFLOW included
    ├── utils/
    │   ├── __init__.py
    │   └── config.py
    └── __init__.py
├── Dockerfile
├── README.md
├── render.yaml
├── .env (MAKE SURE TO ADD IN GITIGNORE FILE)
├── .gitattributes
├── requirements.txt
├── run_pipeline.py # if user prefers web version by python script
└── shellscript.sh  # if user prefers web version by shell script

---

1. Create Virtual Environment

python -m venv venv

2. Activate Virtual Environment

Windows:

venv\Scripts\activate

Mac/Linux:

source venv/bin/activate

---

3. Install Requirements

pip install -r requirements.txt
pip install -r requirements-dev.txt

---

OpenCV Version (Desktop Webcam)

1. Preprocess Data

python -m src.data.preprocess

2. Train Model

python -m src.models.train

3. Run Webcam Detection

python -m src.inference.webcam