A Machine Learning Pipeline for Deepfake Detection

Abstract

Deepfake detection methods need better benchmarks- Accuracy, cost, time. We are doing that via DDP and Juneberry:- Data collection- Data transformation- Modeling- Evaluation Preliminary results confirm that generalizability is a problem. - We expect to improve models with ensemble detectors via DDP.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 14, 2022
Accession Number
AD1183635

Entities

People

  • Catherine Bernaciak
  • Dominic Ross
  • Jeffrey Mellon
  • Shannon Gallagher

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Data Sets
  • Deep Learning
  • Deepfakes
  • Department Of Defense
  • Detection
  • Detectors
  • Engineering
  • Governments
  • Guarantees
  • Images
  • Learning
  • Machine Learning
  • Materials
  • Neural Networks
  • Pipelines
  • Software Development
  • Universities
  • Video

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Parallel and Distributed Computing.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks