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

Recent work in adversarial machine learning started to focus on the visualperception in autonomous driving and studied Adversarial Examples (AEs) forobject detection models. However, in such visual perception pipeline thedetected objects must also be tracked, in a process called Multiple ObjectTracking (MOT), to build the moving trajectories of surrounding obstacles.Since MOT is designed to be robust against errors in object detection, it posesa general challenge to existing attack techniques that blindly target objectiondetection: we find that a success rate of over 98% is needed for them toactually affect the tracking results, a requirement that no existing attacktechnique can satisfy. In this paper, we are the first to study adversarialmachine learning attacks against the complete visual perception pipeline inautonomous driving, and discover a novel attack technique, tracker hijacking,that can effectively fool MOT using AEs on object detection. Using ourtechnique, successful AEs on as few as one single frame can move an existingobject in to or out of the headway of an autonomous vehicle to cause potentialsafety hazards. We perform evaluation using the Berkeley Deep Drive dataset andfind that on average when 3 frames are attacked, our attack can have a nearly100% success rate while attacks that blindly target object detection only haveup to 25%.

Fooling Detection Alone is Not Enough: First Adversarial Attack against Multiple Object Tracking