This repository contains a deep learning-based model for detecting whether an image is a deep fake or not. The model analyzes facial features and artifacts commonly found in deep fake images and returns a classification result.
Detects deep fake images with high accuracy. Pre-trained on a large dataset of real and fake images. Lightweight and easy to integrate into existing projects.
Before setting up the project, ensure you have the following dependencies installed:
- Python 3.7+
- TensorFlow or PyTorch (depending on the model)
- OpenCV
- NumPy
- Matplotlib
- gradio
- Pillow
- facenet-pytorch
- torch
- opencv-python
- grad-cam
Download the pre-trained model from Hugging Face.
Visit the Hugging Face Model Hub and download the files of examples.zip and resnetinceptionv1_epoch_32.pth from the files there.
Run the script to detect whether the image is a deep fake:
Contributions are welcome! Please fork the repository, create a new branch, and submit a pull request.
This project is licensed under the MIT License. See the LICENSE file for more information.
Hugging Face for providing pre-trained models. The open-source community for creating and maintaining valuable tools.