Evolution of Detection Performance Throughout the Online Lifespan of Synthetic Images

  • Dimitrios Karageogiou
  • , Quentin Bammey
  • , Valentin Porcellini
  • , Bertrand Goupil
  • , Denis Teyssou
  • , Symeon Papadopoulos

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Synthetic images disseminated online significantly differ from those used during the training and evaluation of the state-of-the-art detectors. In this work, we analyze the performance of synthetic image detectors as deceptive synthetic images evolve throughout their online lifespan. Our study reveals that, despite advancements in the field, current state-of-the-art detectors struggle to distinguish between synthetic and real images in the wild. Moreover, we show that the time elapsed since the initial online appearance of a synthetic image negatively affects the performance of most detectors. Ultimately, by employing a retrieval-assisted detection approach, we demonstrate the feasibility to maintain initial detection performance throughout the whole online lifespan of an image and enhance the average detection efficacy across several state-of-the-art detectors by 6.7% and 7.8% for balanced accuracy and AUC metrics, respectively.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2024 Workshops, Proceedings
EditorsAlessio Del Bue, Cristian Canton, Jordi Pont-Tuset, Tatiana Tommasi
PublisherSpringer Science and Business Media Deutschland GmbH
Pages400-417
Number of pages18
ISBN (Print)9783031926471
DOIs
Publication statusPublished - 1 Jan 2025
Externally publishedYes
EventWorkshops that were held in conjunction with the 18th European Conference on Computer Vision, ECCV 2024 - Milan, Italy
Duration: 29 Sept 20244 Oct 2024

Publication series

NameLecture Notes in Computer Science
Volume15643 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceWorkshops that were held in conjunction with the 18th European Conference on Computer Vision, ECCV 2024
Country/TerritoryItaly
CityMilan
Period29/09/244/10/24

Fingerprint

Dive into the research topics of 'Evolution of Detection Performance Throughout the Online Lifespan of Synthetic Images'. Together they form a unique fingerprint.

Cite this