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

Modern deepfake detectors have achieved encouraging results, when trainingand test images are drawn from the same data collection. However, when thesedetectors are applied to images produced with unknown deepfake-generationtechniques, considerable performance degradations are commonly observed. Inthis paper, we propose a novel deepfake detector, called SeeABLE, thatformalizes the detection problem as a (one-class) out-of-distribution detectiontask and generalizes better to unseen deepfakes. Specifically, SeeABLE firstgenerates local image perturbations (referred to as soft-discrepancies) andthen pushes the perturbed faces towards predefined prototypes using a novelregression-based bounded contrastive loss. To strengthen the generalizationperformance of SeeABLE to unknown deepfake types, we generate a rich set ofsoft discrepancies and train the detector: (i) to localize, which part of theface was modified, and (ii) to identify the alteration type. To demonstrate thecapabilities of SeeABLE, we perform rigorous experiments on several widely-useddeepfake datasets and show that our model convincingly outperforms competingstate-of-the-art detectors, while exhibiting highly encouraging generalizationcapabilities.

SeeABLE: Soft Discrepancies and Bounded Contrastive Learning for Exposing Deepfakes