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

With the rapid development of generation model, AI-based face manipulationtechnology, which called DeepFakes, has become more and more realistic. Thismeans of face forgery can attack any target, which poses a new threat topersonal privacy and property security. Moreover, the misuse of synthetic videoshows potential dangers in many areas, such as identity harassment, pornographyand news rumors. Inspired by the fact that the spatial coherence and temporalconsistency of physiological signal are destroyed in the generated content, weattempt to find inconsistent patterns that can distinguish between real videosand synthetic videos from the variations of facial pixels, which are highlyrelated to physiological information. Our approach first applies Eulerian VideoMagnification (EVM) at multiple Gaussian scales to the original video toenlarge the physiological variations caused by the change of facial bloodvolume, and then transform the original video and magnified videos into aMulti-Scale Eulerian Magnified Spatial-Temporal map (MEMSTmap), which canrepresent time-varying physiological enhancement sequences on differentoctaves. Then, these maps are reshaped into frame patches in column units andsent to the vision Transformer to learn the spatio-time descriptors of framelevels. Finally, we sort out the feature embedding and output the probabilityof judging whether the video is real or fake. We validate our method on theFaceForensics++ and DeepFake Detection datasets. The results show that ourmodel achieves excellent performance in forgery detection, and also showoutstanding generalization capability in cross-data domain.

FakeTransformer: Exposing Face Forgery From Spatial-Temporal Representation Modeled By Facial Pixel Variations