Deep Fake Image Classifier 🕵️‍♂️

A Streamlit web app that uses a TensorFlow Lite model to classify images as either Real or Fake. This project helps detect deep fake images using deep learning pretrained model optimized for lightweight performance.

---

🗕️ Features

• Upload an image and detect whether it's Real or Fake.
• Displays the confidence score of the prediction.
• Lightweight and fast inference with TensorFlow Lite.
• Simple and user-friendly Streamlit interface.

---

🛠️ Installation

1. Clone the repository

git clone https://github.com/AhmedFoda54/Deep-Fake-Classification.git

2. Create and activate a virtual environment (optional but recommended)

python -m venv venv
source venv/bin/activate  # On Windows use: venv\Scripts\activate

3. Install the required packages

pip install -r requirements.txt

---

🚀 Running the App

1. Run the Streamlit app:

streamlit run app/main.py

2. Open your browser and navigate to the URL provided in the terminal (usually http://localhost:8501).

---

📂 Project Structure

deepfake-classifier/
  |--app/
        |-- main.py                      # Main Streamlit application file
        |-- model_utils.py                      # Helper Functions
  |--model_convertion/
        |-- convert_to_tflite.py  # To convert the .h5/keras model to .tflite model
  |--deepfake_classifier.tflite # TensorFlow Lite model file (downloaded automatically)
  |-- requirements.txt            # List of required Python packages
  |-- README.md                   # Project documentation

---

🎯 Model Information

• The model is hosted on Dropbox and automatically downloaded when you run the app.
• It uses a TensorFlow Lite format optimized for fast, efficient performance.

---

🌐 How to Use

1. Upload an Image: Click on the file uploader and select an image (.jpg, .jpeg, or .png).
2. View the Image: The app displays the uploaded image.
3. Click Predict: Press the "Predict" button to classify the image.
4. View Results: The app will show whether the image is Real or Fake, along with a confidence score.

---

📊 Dependencies

Make sure the following packages are installed (included in requirements.txt):

• streamlit
• tensorflow
• numpy
• pillow
• gdown
• os

---

💼 Author

• Ahmed Foda
  Data Science & Coding Instructor Enthusiast
  LinkedIn | Email

---

💍 License

This project is licensed under the MIT License.

---

🔗 References

1. Heidari, A., Navimipour, N. J., Dag, H., & Unal, M. (2023). Deepfake detection using deep learning methods: A systematic and comprehensive review. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 14(2). https://doi.org/10.1002/widm.1520

2. Abir, W. H., et al. (2022). Detecting deepfake images using deep learning techniques and explainable AI methods. Intelligent Automation & Soft Computing, 35(2), 2151–2169. https://doi.org/10.32604/iasc.2023.029653

3. Salih, A. M., Raisi-Estabragh, Z., Galazzo, I. B., Radeva, P., Petersen, S. E., Lekadir, K., & Menegaz, G. (2024). A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME. arXiv. https://arxiv.org/abs/2305.02012