Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 557 Bytes

File metadata and controls

9 lines (6 loc) · 557 Bytes

DeepFake-Classification

We are confronted with an unprecedented risk of gross violations of basic human rights, as well as a fundamental, unavoidable shift in how humans interact socially. We've previously seen examples of defamation and manipulation of news headlines, medical (mis)information, and invasions of privacy. The purpose of this proposed project is to detect DeepFake photos using an internet image database.

We aimed to compare three different convolutional neural networks:

  1. Custom CNN Architectures.
  2. VGGFace16
  3. DenseNet-121