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

Adversarial attack arises due to the vulnerability of deep neural networks toperceive input samples injected with imperceptible perturbations. Recently,adversarial attack has been applied to visual object tracking to evaluate therobustness of deep trackers. Assuming that the model structures of deeptrackers are known, a variety of white-box attack approaches to visual trackinghave demonstrated promising results. However, the model knowledge about deeptrackers is usually unavailable in real applications. In this paper, we proposea decision-based black-box attack method for visual object tracking. Incontrast to existing black-box adversarial attack methods that deal with staticimages for image classification, we propose IoU attack that sequentiallygenerates perturbations based on the predicted IoU scores from both current andhistorical frames. By decreasing the IoU scores, the proposed attack methoddegrades the accuracy of temporal coherent bounding boxes (i.e., objectmotions) accordingly. In addition, we transfer the learned perturbations to thenext few frames to initialize temporal motion attack. We validate the proposedIoU attack on state-of-the-art deep trackers (i.e., detection based,correlation filter based, and long-term trackers). Extensive experiments on thebenchmark datasets indicate the effectiveness of the proposed IoU attackmethod. The source code is available athttps://github.com/VISION-SJTU/IoUattack.

IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking