🕵️ Detect Deepfakes: Sense

Welcome to Sense’s open source repository

Secure your identity verification systems with Sense’s powerful deepfake detection. Prevent spoofing, fraud, and identity theft using advanced machine learning and computer vision

🧩 Overview

As digital identity becomes the cornerstone of secure online interactions, the threat landscape has evolved far beyond password breaches and phishing.

Modern attackers now leverage advanced spoofing techniques like a GAN image which is a synthetic (i.e., fake) image produced by an AI model trained to generate realistic-looking images that resemble real ones, such as human faces—to trick facial recognition systems.

This calls for a new generation of AI-powered defenses that go beyond static image verification and ensure the authenticity of every face presented in digital workflows.

🔧 Features

1. ONNX Runtime for inference
2. Image upload and classification endpoint (/deepfake)
3. Cross-Origin support (CORS) for frontend integration
4. Dockerized for easy deployment
5. Model visualization support

⏳ Clone the Repository

# Clone the repository
git clone https://github.com/sense-opensource/sense-deepfake-detector.git

# Navigate into the project directory
cd sense-deepfake-detector

🧠 Model

The ONNX model file is not included in the repository. You must download the model file manually or programmatically and place it in the appropriate folder.

✅ Download Instructions

Download the model file from the below link:

[Model] (https://github.com/sense-opensource/sense-deepfake-detector/releases/download/v1.0.0/efficientnet-b7.onnx this file needs to be placed inside the models folder)

Ensure the model is saved in:  models/efficientnet-b7.onnx

2. API Configuration

Method 1: Install Python Dependencies

pip install -r requirements.txt

Start the FastAPI Server

uvicorn app:app --reload

This will start the API server on: http://localhost:3015

Method 2: Running with Docker

Build Docker Image

docker build -t sense_deepfake_opensource_image .

Run Docker Container

docker run -d --name sense_deepfake_opensource_container -p 3015:3015 sense_deepfake_opensource_image

This will start the API server on: http://localhost:3015

3. Run the Frontend

cd front-end
npm install
npm run dev

By default, the frontend runs on: http://localhost:5000

Useful Docker Commands

Stop container

docker stop sense_deepfake_opensource_container

Remove container

docker rm -f sense_deepfake_opensource_container

Stop container

docker stop sense_deepfake_opensource_container

Remove image

docker rmi -f sense_deepfake_opensource_image

View logs

docker logs sense_deepfake_opensource_container

MIT License — free to use, share, and modify