Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DeepFake-O-Meter v2.0: An Open Platform for DeepFake Detection

Yan Ju, Chengzhe Sun, Shan Jia, Shuwei Hou, Zhaofeng Si, Soumyya Kanti Datta, Lipeng Ke, Riky Zhou, Anita Nikolich, Siwei Lyu

TL;DR

The paper introduces DeepFake-O-Meter v2.0, an open, multi-modal platform for detecting DeepFakes in images, videos, and audio. It combines a front-end UI and a GPU-backed back-end that containerizes detector methods, enabling scalable, parallel evaluation and fair task scheduling via a job-balancing mechanism. Key contributions include a redesigned user interface with tiered access, a Docker-based detector repository (covering diverse image, video, and audio detectors), and a comprehensive usage-analysis framework that tracks user activity and detector performance. The platform serves both public users and digital-forensics researchers, enabling practical DeepFake analysis, benchmarking, and data-driven improvement of detection methods. Future work envisions integrating additional detectors, multi-modal approaches, and deeper performance evaluations on real-world data to guide detector selection on the platform.

Abstract

Deepfakes, as AI-generated media, have increasingly threatened media integrity and personal privacy with realistic yet fake digital content. In this work, we introduce an open-source and user-friendly online platform, DeepFake-O-Meter v2.0, that integrates state-of-the-art methods for detecting Deepfake images, videos, and audio. Built upon DeepFake-O-Meter v1.0, we have made significant upgrades and improvements in platform architecture design, including user interaction, detector integration, job balancing, and security management. The platform aims to offer everyday users a convenient service for analyzing DeepFake media using multiple state-of-the-art detection algorithms. It ensures secure and private delivery of the analysis results. Furthermore, it serves as an evaluation and benchmarking platform for researchers in digital media forensics to compare the performance of multiple algorithms on the same input. We have also conducted detailed usage analysis based on the collected data to gain deeper insights into our platform's statistics. This involves analyzing two-month trends in user activity and evaluating the processing efficiency of each detector.

DeepFake-O-Meter v2.0: An Open Platform for DeepFake Detection

TL;DR

The paper introduces DeepFake-O-Meter v2.0, an open, multi-modal platform for detecting DeepFakes in images, videos, and audio. It combines a front-end UI and a GPU-backed back-end that containerizes detector methods, enabling scalable, parallel evaluation and fair task scheduling via a job-balancing mechanism. Key contributions include a redesigned user interface with tiered access, a Docker-based detector repository (covering diverse image, video, and audio detectors), and a comprehensive usage-analysis framework that tracks user activity and detector performance. The platform serves both public users and digital-forensics researchers, enabling practical DeepFake analysis, benchmarking, and data-driven improvement of detection methods. Future work envisions integrating additional detectors, multi-modal approaches, and deeper performance evaluations on real-world data to guide detector selection on the platform.

Abstract

Deepfakes, as AI-generated media, have increasingly threatened media integrity and personal privacy with realistic yet fake digital content. In this work, we introduce an open-source and user-friendly online platform, DeepFake-O-Meter v2.0, that integrates state-of-the-art methods for detecting Deepfake images, videos, and audio. Built upon DeepFake-O-Meter v1.0, we have made significant upgrades and improvements in platform architecture design, including user interaction, detector integration, job balancing, and security management. The platform aims to offer everyday users a convenient service for analyzing DeepFake media using multiple state-of-the-art detection algorithms. It ensures secure and private delivery of the analysis results. Furthermore, it serves as an evaluation and benchmarking platform for researchers in digital media forensics to compare the performance of multiple algorithms on the same input. We have also conducted detailed usage analysis based on the collected data to gain deeper insights into our platform's statistics. This involves analyzing two-month trends in user activity and evaluating the processing efficiency of each detector.
Paper Structure (15 sections, 6 figures, 3 tables)

This paper contains 15 sections, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Platform Design
  3. Front-end Design
  4. Account System
  5. Task Submission
  6. Result Display
  7. Submission History
  8. Back-end Design
  9. Overview
  10. Container Building
  11. Job Balancing
  12. Platform Data Analysis
  13. Usage Analysis
  14. Detector Analysis
  15. Conclusion

Figures (6)

  • Figure 1: DeepFake-O-Meter web pages, (a) Homepage, (b) Result Page, (c) Detailed Result Page, (d) Submission History Page.
  • Figure 2: Pipeline of the job balancing module.
  • Figure 3: User pattern statistics based on data from February 8, 2024, to June 1, 2024.
  • Figure 4: Accumulative daily submission activity of our platform.
  • Figure 5: Running time of each module
  • ...and 1 more figures