Skip to content

JARVVVIS/DeepFake-Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

DeepFake Detection

DeepFakes have been in the news for over a year now, mostly for all the wrong reasons. This repository is home to Pytorch implementations of various deepfake detection algorithms, tested on Google AI's Deep Fake Detection Dataset.

Algorithms

  1. Data : Due to computational constraints, trained the MesoInception4 network on a subset of the DeepFake Detection dataset(363 videos from both real and manipulated classes).

    Results : The data was trained for 30 epochs using all default hyperparamametes and network as mentioned in the original paper.

     Training Accuracy: 93.75% 
     Test Accuracy: 90.42%
    

About

Various DeepFake techniques implemented in Pytorch

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published