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 facial forgery techniques, forgery detectionhas attracted more and more attention due to security concerns. Existingapproaches attempt to use frequency information to mine subtle artifacts underhigh-quality forged faces. However, the exploitation of frequency informationis coarse-grained, and more importantly, their vanilla learning processstruggles to extract fine-grained forgery traces. To address this issue, wepropose a progressive enhancement learning framework to exploit both the RGBand fine-grained frequency clues. Specifically, we perform a fine-graineddecomposition of RGB images to completely decouple the real and fake traces inthe frequency space. Subsequently, we propose a progressive enhancementlearning framework based on a two-branch network, combined withself-enhancement and mutual-enhancement modules. The self-enhancement modulecaptures the traces in different input spaces based on spatial noiseenhancement and channel attention. The Mutual-enhancement module concurrentlyenhances RGB and frequency features by communicating in the shared spatialdimension. The progressive enhancement process facilitates the learning ofdiscriminative features with fine-grained face forgery clues. Extensiveexperiments on several datasets show that our method outperforms thestate-of-the-art face forgery detection methods.

Exploiting Fine-grained Face Forgery Clues via Progressive Enhancement Learning