Skip to content

A deep learning-based web application for deepfake video detection, powered by the fine-tuned XceptionNet (Extreme Inception) model. The system allows users to upload videos for deepfake detection, processes them through the trained model, and provides results via a clean Django-based web interface.

License

Notifications You must be signed in to change notification settings

maheenshkk/deepfake-detection-system-deepdeception

Repository files navigation

DeepDeception

A Deep Learning-based web application for deepfake video detection, powered by the Xception model and deployed via Railway.


Overview

DeepDeception uses a fine-tuned Xception (Extreme Inception) model to detect deepfakes from video uploads. The system is integrated into a Django-based web app with a clean user interface, a feedback submission form, and a backend database to store user feedback and email addresses.


Core Features

  • Xception-based Deepfake Detection
    Trained and tuned on real vs fake video data (DFDC Dataset)

  • Django Web App Interface
    Upload video files and get instant predictions

  • Feedback System
    Users can submit feedback and emails, which are stored in a secure database

  • Deployed on Railway
    Accessible online with scalable backend hosting


Tech Stack

Component Technology
Deep Learning Xception (Keras/TensorFlow)
Web Framework Django (Python)
Frontend HTML, CSS, Bootstrap
Database SQLite / PostgreSQL (Railway)
Deployment Railway
Video Handling OpenCV, FFmpeg

Project Structure

DeepDeception/ │

├── model/ # Xception model + inference code

├── webapp/ # Django project

│ ├── templates/

│ ├── static/

│ ├── views.py

│ ├── urls.py

│ └── models.py # Stores email + feedback

├── media/ # Uploaded videos (temp storage)

├── requirements.txt

└── README.md

Preview

image image image

How to Run Locally

  1. Clone the repository:
git clone https://github.com/YourUsername/DeepDeception.git
cd DeepDeception

pip install -r requirements.txt

python manage.py runserver
 Feedback System
Users can submit feedback through the web interface.

Submitted feedback and email addresses are stored securely in the backend database.

Admins can view and manage feedback entries via the Django admin panel (if enabled).

 Live Demo
 Hosted on Railway
URL: [Insert your deployed Railway app link here]

 Model Performance
Architecture: Xception (pretrained, fine-tuned)

Dataset: Deepfake Detection Challenge (DFDC)

Accuracy: ~94% on validation set

Input: Short video clips (MP4)

Output: Real / Fake classification with confidence score

 Authors
Maheen Shaikh
email: maheenshkk@gmail.com 
[@maheenshkk](https://github.com/maheenshkk)
AI/ML Engineer | Data Analytics | Computer Vision | NLP | Power BI | Python

Abdulrehman Qureshi
[crazy-scientistt](https://github.com/crazy-scientistt)
AI Researcher | Computer Vision & Web Systems




About

A deep learning-based web application for deepfake video detection, powered by the fine-tuned XceptionNet (Extreme Inception) model. The system allows users to upload videos for deepfake detection, processes them through the trained model, and provides results via a clean Django-based web interface.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 2

  •  
  •