SEI-CMU Deepfakes Work

Abstract

A deep neural network (DNN) is a neural network with many layers used to convert inputs to targets. A deepfake is a media file of a human subject that has been manipulated using a DNN. Deepfakes are more complex than other digital forgeries. It is difficult, but not impossible to make high-quality deepfakes with enough time, effort, skill, and resources. Given the potential scale and complexity of deepfakes, automatic detection methods are needed. It is difficult, but not impossible to detect deepfakes automatically However, detection methods must continue to evolve to maintain upper-hand.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 16, 2023
Accession Number
AD1204246

Entities

People

  • Catherine Bernaciak
  • Dominic Ross
  • Matthew E. Walsh

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Cardiac Arrhythmias
  • Computer Programming
  • Computers
  • Deepfakes
  • Department Of Defense
  • Detection
  • Detectors
  • Disinformation
  • Education
  • Engineering
  • Facial Recognition
  • Frequency
  • Frequency Domain
  • Guarantees
  • Heart Rate
  • Images
  • Light Sources
  • Machine Learning
  • Materials
  • Neural Networks
  • Open Source Software
  • Software Development
  • Universities

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Vision.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks