English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Images synthesized by powerful generative adversarial network (GAN) basedmethods have drawn moral and privacy concerns. Although image forensic modelshave reached great performance in detecting fake images from real ones, thesemodels can be easily fooled with a simple adversarial attack. But, the noiseadding adversarial samples are also arousing suspicion. In this paper, insteadof adding adversarial noise, we optimally search adversarial points on facemanifold to generate anti-forensic fake face images. We iteratively do agradient-descent with each small step in the latent space of a generativemodel, e.g. Style-GAN, to find an adversarial latent vector, which is similarto norm-based adversarial attack but in latent space. Then, the generated fakeimages driven by the adversarial latent vectors with the help of GANs candefeat main-stream forensic models. For examples, they make the accuracy ofdeepfake detection models based on Xception or EfficientNet drop from over 90%to nearly 0%, meanwhile maintaining high visual quality. In addition, we findmanipulating style vector $z$ or noise vectors $n$ at different levels haveimpacts on attack success rate. The generated adversarial images mainly havefacial texture or face attributes changing.

Exploring Adversarial Fake Images on Face Manifold