This repository contains the code and resources for my Science project: Deepface Image Detection. The goal of this project is describe in this abstract below.
The "Deepface Image Detection" project focuses on the development of neural networks for identifying deepfake images, leveraging Convolutional Neural Networks (CNN) and the InceptionV3 architecture. The model achieved exceptional performance, exhibiting a remarkable accuracy of over 99% on the training set and exceeding 95% on the validation set.
The dataset employed in this project is the Deepface HQ dataset, as detailed in the research paper titled "Unmasking DeepFakes with simple Features - Ricard Durall, Margret Keuper, Franz-Josef Pfreundt, Janis Keuper" accessible at https://arxiv.org/abs/1911.00686.
Through this project, we aim to contribute to the ongoing efforts in combating the proliferation of deceptive visual content online. Researchers, developers, and interested parties can utilize this open-source repository to access the trained model, explore the dataset, and extend the work to address real-world challenges posed by deepfake images.
To use this project, you'll need to clone the repository and install the required dependencies.
git clone https://github.com/your_username/your_project.git
cd your_project
pip install -r requirements.txt
The project hierachy should be construct like this for direct running
The dataset can be downloaded in https://drive.google.com/drive/folders/1_q8RhK9PNQTyIp5fsaFkxVfsbngBafRT?usp=sharing
The accuracy logs of InceptionV3 model
The accuracy logs of CNN model
Thank you for your interest in our "Deepface Image Detection" project. If you have any questions, suggestions, or feedback, please feel free to reach out to us.
Project Maintainer: [Hoang HaDang]
Email: [danghoang2109@gmail.com] [hoanghd.17@grad.uit.edu.vn]
GitHub: [github.com/DangHoang2109]
Linkedin: [Hoàng Đăng]
We appreciate your support and look forward to hearing from you!