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

We propose a stealthy clean-label video backdoor attack against Deep Learning(DL)-based models aiming at detecting a particular class of spoofing attacks,namely video rebroadcast attacks. The injected backdoor does not affectspoofing detection in normal conditions, but induces a misclassification in thepresence of a specific triggering signal. The proposed backdoor relies on atemporal trigger altering the average chrominance of the video sequence. Thebackdoor signal is designed by taking into account the peculiarities of theHuman Visual System (HVS) to reduce the visibility of the trigger, thusincreasing the stealthiness of the backdoor. To force the network to look atthe presence of the trigger in the challenging clean-label scenario, we choosethe poisoned samples used for the injection of the backdoor following aso-called Outlier Poisoning Strategy (OPS). According to OPS, the triggeringsignal is inserted in the training samples that the network finds moredifficult to classify. The effectiveness of the proposed backdoor attack andits generality are validated experimentally on different datasets andanti-spoofing rebroadcast detection architectures.

A temporal chrominance trigger for clean-label backdoor attack against anti-spoof rebroadcast detection