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

The rapid progression of Generative Adversarial Networks (GANs) has raised aconcern of their misuse for malicious purposes, especially in creating fakeface images. Although many proposed methods succeed in detecting GAN-basedsynthetic images, they are still limited by the need for large quantities ofthe training fake image dataset and challenges for the detector'sgeneralizability to unknown facial images. In this paper, we propose a newapproach that explores the asynchronous frequency spectra of color channels,which is simple but effective for training both unsupervised and supervisedlearning models to distinguish GAN-based synthetic images. We furtherinvestigate the transferability of a training model that learns from oursuggested features in one source domain and validates on another target domainswith prior knowledge of the features' distribution. Our experimental resultsshow that the discrepancy of spectra in the frequency domain is a practicalartifact to effectively detect various types of GAN-based generated images.

Exploring the Asynchronous of the Frequency Spectra of GAN-generated Facial Images