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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

As video analysis using deep learning models becomes more widespread, thevulnerability of such models to adversarial attacks is becoming a pressingconcern. In particular, Universal Adversarial Perturbation (UAP) poses asignificant threat, as a single perturbation can mislead deep learning modelson entire datasets. We propose a novel video UAP using image data and imagemodel. This enables us to take advantage of the rich image data and imagemodel-based studies available for video applications. However, there is achallenge that image models are limited in their ability to analyze thetemporal aspects of videos, which is crucial for a successful video attack. Toaddress this challenge, we introduce the Breaking Temporal Consistency (BTC)method, which is the first attempt to incorporate temporal information intovideo attacks using image models. We aim to generate adversarial videos thathave opposite patterns to the original. Specifically, BTC-UAP minimizes thefeature similarity between neighboring frames in videos. Our approach is simplebut effective at attacking unseen video models. Additionally, it is applicableto videos of varying lengths and invariant to temporal shifts. Our approachsurpasses existing methods in terms of effectiveness on various datasets,including ImageNet, UCF-101, and Kinetics-400.

Breaking Temporal Consistency: Generating Video Universal Adversarial Perturbations Using Image Models