README.md

metadata

title: DeepFake Detection Demo
emoji: 🧠
colorFrom: indigo
colorTo: purple
sdk: streamlit
sdk_version: 1.37.1
app_file: app.py
pinned: false

🧠 DeepFake Detection Demo

This application demonstrates a CPU-based deepfake detection model for deep fake rejection including Style GAN and Diffusion model architectures

🔍 What This Demo Does

• Accepts an uploaded face image
• Classifies it as Real or Fake
• Displays confidence and probabilities
• Runs entirely on CPU (TorchScript)

🏗 Architecture

• Model: Private Hugging Face Model Repository
• App: Hugging Face Space (Streamlit)
• Runtime: CPU (no GPU required)

⚠️ Further Inquiries

Please feel free to contact uzair.mughal@unikrew.com for any queries or information

📌 Known Limitations

• Further releases will enhance the model for face editing using AI (such as the swapping of a real face on the body of a real person -- much like a photoshop but using AI)