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

3D deep models consuming point clouds have achieved sound application effectsin computer vision. However, recent studies have shown they are vulnerable to3D adversarial point clouds. In this paper, we regard these malicious pointclouds as 3D steganography examples and present a new perspective, 3Dsteganalysis, to counter such examples. Specifically, we propose 3D-VFD, avictim-free detector against 3D adversarial point clouds. Its core idea is tocapture the discrepancies between residual geometric feature distributions ofbenign point clouds and adversarial point clouds and map these point clouds toa lower dimensional space where we can efficiently distinguish them. Unlikeexisting detection techniques against 3D adversarial point clouds, 3D-VFD doesnot rely on the victim 3D deep model's outputs for discrimination. Extensiveexperiments demonstrate that 3D-VFD achieves state-of-the-art detection and caneffectively detect 3D adversarial attacks based on point adding and pointperturbation while keeping fast detection speed.

3D-VFD: A Victim-free Detector against 3D Adversarial Point Clouds