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

In this paper we propose a novel human-centered approach for detectingforgery in face images, using dynamic prototypes as a form of visualexplanations. Currently, most state-of-the-art deepfake detections are based onblack-box models that process videos frame-by-frame for inference, and fewclosely examine their temporal inconsistencies. However, the existence of suchtemporal artifacts within deepfake videos is key in detecting and explainingdeepfakes to a supervising human. To this end, we propose Dynamic PrototypeNetwork (DPNet) -- an interpretable and effective solution that utilizesdynamic representations (i.e., prototypes) to explain deepfake temporalartifacts. Extensive experimental results show that DPNet achieves competitivepredictive performance, even on unseen testing datasets such as Google'sDeepFakeDetection, DeeperForensics, and Celeb-DF, while providing easyreferential explanations of deepfake dynamics. On top of DPNet's prototypicalframework, we further formulate temporal logic specifications based on thesedynamics to check our model's compliance to desired temporal behaviors, henceproviding trustworthiness for such critical detection systems.

Interpretable and Trustworthy Deepfake Detection via Dynamic Prototypes