Detects whether an image is real or deepfake using Transformers
Project Description
This project focuses on detecting deepfake images by analyzing visual patterns using advanced deep learning techniques. A Vision Transformer–based model is used to classify images as real or manipulated. The system provides an easy-to-use web interface where users can upload images and instantly get prediction results, making it suitable for real-world and academic use cases.
Features
Detects real vs deepfake images with high accuracy
Uses Vision Transformer (ViT) for image classification
Image upload–based prediction through web interface
Clean and well-documented source code
Suitable for academic and final-year projects
Tech Stack
Python
Hugging Face (Pre-trained Models)
Vision Transformer (ViT)
Flask (Backend Framework)
HTML, JavaScript (Frontend)
NumPy, Pandas, and other essential Python libraries
System Requirements
Python 3.10+
Anaconda
No GPU required
Deliverables
Jupyter Notebook
Dataset
Setup over video call
WhatsApp support
WhatsApp For Project