Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381547719> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4381547719 endingPage "106594" @default.
- W4381547719 startingPage "106594" @default.
- W4381547719 abstract "The emergence of visually realistic GAN-generated facial images has raised concerns regarding potential misuse. In response, effective forensic algorithms have been developed to detect such synthetic images in recent years. However, the vulnerability of such forensic detectors to adversarial attacks remains an important issue that requires further investigation. In this paper, we propose a new black-box attack method against GAN-generated image detectors. It involves contrastive learning strategy to train an encoder–decoder anti-forensic network with a contrastive loss function. GAN-generated and corresponding simulated real images are constructed as positive and negative samples, respectively. By leveraging the trained attack model, we can apply imperceptible perturbation to input synthetic images for removing GAN fingerprint to some extent. GAN-generated image detectors may be deceived consequently. Extensive experimental results demonstrate that the proposed attack effectively reduces the accuracy of three state-of-the-art detectors on six popular GANs, while also achieving high visual quality of the attacked images. The source code will be available at https://github.com/ZXMMD/BAttGAND." @default.
- W4381547719 created "2023-06-22" @default.
- W4381547719 creator A5024660893 @default.
- W4381547719 creator A5060319992 @default.
- W4381547719 creator A5084481998 @default.
- W4381547719 date "2023-09-01" @default.
- W4381547719 modified "2023-09-27" @default.
- W4381547719 title "Black-box attack against GAN-generated image detector with contrastive perturbation" @default.
- W4381547719 cites W2133665775 @default.
- W4381547719 cites W2180612164 @default.
- W4381547719 cites W2183341477 @default.
- W4381547719 cites W2194775991 @default.
- W4381547719 cites W2531409750 @default.
- W4381547719 cites W2535388113 @default.
- W4381547719 cites W2811414481 @default.
- W4381547719 cites W2913399670 @default.
- W4381547719 cites W2962770929 @default.
- W4381547719 cites W2963767194 @default.
- W4381547719 cites W3034600949 @default.
- W4381547719 cites W3035574324 @default.
- W4381547719 cites W3036426855 @default.
- W4381547719 cites W3040168631 @default.
- W4381547719 cites W3096651815 @default.
- W4381547719 cites W3096831136 @default.
- W4381547719 cites W3170697543 @default.
- W4381547719 cites W3173126908 @default.
- W4381547719 cites W4220744313 @default.
- W4381547719 cites W4320736717 @default.
- W4381547719 doi "https://doi.org/10.1016/j.engappai.2023.106594" @default.
- W4381547719 hasPublicationYear "2023" @default.
- W4381547719 type Work @default.
- W4381547719 citedByCount "0" @default.
- W4381547719 crossrefType "journal-article" @default.
- W4381547719 hasAuthorship W4381547719A5024660893 @default.
- W4381547719 hasAuthorship W4381547719A5060319992 @default.
- W4381547719 hasAuthorship W4381547719A5084481998 @default.
- W4381547719 hasBestOaLocation W43815477192 @default.
- W4381547719 hasConcept C111919701 @default.
- W4381547719 hasConcept C115961682 @default.
- W4381547719 hasConcept C118505674 @default.
- W4381547719 hasConcept C153180895 @default.
- W4381547719 hasConcept C154945302 @default.
- W4381547719 hasConcept C177264268 @default.
- W4381547719 hasConcept C199360897 @default.
- W4381547719 hasConcept C2776760102 @default.
- W4381547719 hasConcept C31972630 @default.
- W4381547719 hasConcept C41008148 @default.
- W4381547719 hasConcept C76155785 @default.
- W4381547719 hasConcept C94915269 @default.
- W4381547719 hasConceptScore W4381547719C111919701 @default.
- W4381547719 hasConceptScore W4381547719C115961682 @default.
- W4381547719 hasConceptScore W4381547719C118505674 @default.
- W4381547719 hasConceptScore W4381547719C153180895 @default.
- W4381547719 hasConceptScore W4381547719C154945302 @default.
- W4381547719 hasConceptScore W4381547719C177264268 @default.
- W4381547719 hasConceptScore W4381547719C199360897 @default.
- W4381547719 hasConceptScore W4381547719C2776760102 @default.
- W4381547719 hasConceptScore W4381547719C31972630 @default.
- W4381547719 hasConceptScore W4381547719C41008148 @default.
- W4381547719 hasConceptScore W4381547719C76155785 @default.
- W4381547719 hasConceptScore W4381547719C94915269 @default.
- W4381547719 hasFunder F4320321001 @default.
- W4381547719 hasFunder F4320335787 @default.
- W4381547719 hasLocation W43815477191 @default.
- W4381547719 hasLocation W43815477192 @default.
- W4381547719 hasOpenAccess W4381547719 @default.
- W4381547719 hasPrimaryLocation W43815477191 @default.
- W4381547719 hasRelatedWork W1891287906 @default.
- W4381547719 hasRelatedWork W1969923398 @default.
- W4381547719 hasRelatedWork W2036807459 @default.
- W4381547719 hasRelatedWork W2130228941 @default.
- W4381547719 hasRelatedWork W2161229648 @default.
- W4381547719 hasRelatedWork W2166024367 @default.
- W4381547719 hasRelatedWork W2275988210 @default.
- W4381547719 hasRelatedWork W2772917594 @default.
- W4381547719 hasRelatedWork W2775347418 @default.
- W4381547719 hasRelatedWork W2993674027 @default.
- W4381547719 hasVolume "124" @default.
- W4381547719 isParatext "false" @default.
- W4381547719 isRetracted "false" @default.
- W4381547719 workType "article" @default.