Skip to content

SahanaAvantika/hackhers

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Hackhers

HackHers is a website designed to combat the rise of manipulated media, including deepfakes and AI-generated videos. By enabling users to verify the authenticity of uploaded videos, it helps creators protect their work from misuse and empowers viewers to make informed judgments.

In addition to detecting manipulation, HackHers safeguards personal privacy by alerting users to content that may be deceptively altered, preventing the spread of misinformation and protecting individuals from impersonation or identity exploitation.


Features

  • 📱 Responsive Website: Seamlessly optimized for both desktop and mobile.
  • 🧠 AI-Powered Detection: Combines on-device and cloud-based ML models to identify subtle signs of manipulation.
  • 🎥 Frame-by-Frame Video Analysis: Uses OpenCV to break down videos and catch even minor deepfake or synthetic edits.
  • 🚨 Real-Time Results: Instant detection results via Firebase, with clear confidence scores and explanations.
  • 🎓 Educational Insights: Explains manipulation techniques to build digital media literacy.
  • 🔒 Privacy Protection: Identifies misleading or harmful content early, reducing risks of impersonation or identity theft.

Getting Started

Follow these steps to set up and run HackHers locally.

1. Python Version

  • Ensure you have Python 3.9, 3.10, or 3.11 installed.
    ⚠️ Python 3.12 or above may cause compatibility issues.

2. Install the required packages

  • pip install numpy==1.27.5
  • pip install opencv-python
  • pip install tensorflow==2.15.0
  • pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu
  • pip install matplotlib
  • pip install Pillow
  • pip install Flask
  • pip install dlib

3. Run the server

  • cd model
  • python server.py

HackHers was trained on the FaceForensics++ dataset, a widely used benchmark for detecting manipulated videos. This dataset allowed us to fine-tune our models to achieve strong performance in spotting deepfakes and synthetic content.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors