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

Segment Anything Model (SAM) has attracted significant attention recently,due to its impressive performance on various downstream tasks in a zero-shortmanner. Computer vision (CV) area might follow the natural language processing(NLP) area to embark on a path from task-specific vision models towardfoundation models. However, deep vision models are widely recognized asvulnerable to adversarial examples, which fool the model to make wrongpredictions with imperceptible perturbation. Such vulnerability to adversarialattacks causes serious concerns when applying deep models to security-sensitiveapplications. Therefore, it is critical to know whether the vision foundationmodel SAM can also be fooled by adversarial attacks. To the best of ourknowledge, our work is the first of its kind to conduct a comprehensiveinvestigation on how to attack SAM with adversarial examples. With the basicattack goal set to mask removal, we investigate the adversarial robustness ofSAM in the full white-box setting and transfer-based black-box settings. Beyondthe basic goal of mask removal, we further investigate and find that it ispossible to generate any desired mask by the adversarial attack.

Attack-SAM: Towards Attacking Segment Anything Model With Adversarial Examples