Skip to content

MedlynJacob/VisionShield

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

🧠👁️ VisionShield - Hybrid AI System for Real-Time DeepFake Detection

Overview

VisionShield is an ongoing research-driven project exploring the use of deep learning and computer vision techniques to detect AI-generated (DeepFake) media. The system integrates convolutional neural networks (CNNs) with OpenCV-based preprocessing to identify subtle facial artifacts and inconsistencies that distinguish synthetic faces from authentic ones.

Key Objectives

  • Develop a hybrid Neural Network–OpenCV framework capable of detecting DeepFake manipulations in real time.
  • Utilize benchmark datasets such as FaceForensics++ and DeepFake Detection Challenge (DFDC) for model training and evaluation.
  • Apply Grad-CAM visualization to improve explainability by highlighting the facial regions influencing model predictions.
  • Implement a Streamlit-based interface for real-time inference and result interpretation.

Research Focus

This project aims to contribute to the growing field of AI for digital media forensics, emphasizing model interpretability, adversarial robustness, and responsible use of generative AI technologies.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published