Web Deepfake Detect

A web application for detecting deepfake images and videos using machine learning.

Features

• Upload images or videos for deepfake detection
• Batch prediction support
• Model management

Requirements

• Python 3.8+
• pip

Setup

1. Clone the repository

   git clone https://github.com/yourusername/Web_Deepfake_Detect.git
   cd Web_Deepfake_Detect

2. Create and activate a virtual environment

   python -m venv venv
   # On Windows:
   venv\Scripts\activate
   # On macOS/Linux:
   source venv/bin/activate

3. Install dependencies

   pip install -r requirements.txt

4. Download or place your trained model weights

   • Place your model files in the weights/ directory.

Download Pretrained Model Weights

Download the pretrained model weights from this Google Drive folder and place them in the weights/ directory before running the application.

Running the Web Application

1. Start the web server

   python web.py

   or, if using Flask:

   flask run

2. Open your browser

   • Go to http://localhost:5000

3. Upload images or videos

   • Use the web interface to upload files and view predictions.

Batch Prediction

To run predictions on a folder of files and save results to a CSV:

./predict_submission.sh <test_folder> <output.csv>

On Windows, use Git Bash or WSL to run the script.

Notes

• Ensure your model weights are compatible with the code.
• For large files or datasets, adjust the configuration as needed.

License

MIT License

---

Feel free to contribute or open issues!