AI Paper Atlas

Research map / Point clouds / Point cloud classification

Towards a 3D Transfer-based Black-box Attack via Critical Feature Guidance

Shuchao Pang, Zhenghan Chen, Shen Zhang, Liming Lu, Siyuan Liang, Anan Du, Yongbin Zhou

ICCV 2025 · Point clouds › Point cloud classification · approach: Adversarial point cloud attacks · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase perturbation budget epsilon from 0.18 to 0.45 (ablation)CFG · ModelNet40transferability score: 53.1 -> 63.6Improved
increase perturbation budget epsilon from 0.18 to 0.45 (ablation)PF-Attack · ModelNet40transferability score: 40.2 -> 46.8Improved
use CFG instead of PF-AttackPointNet source · ModelNet40transferability score (eps 0.18): 40.2 -> 53.1Improved
use CFG instead of PF-AttackPointNet source · ModelNet40transferability score (eps 0.45): 46.8 -> 63.6Improved
use CFG instead of PF-Attack against SRS defensePointNet source · ModelNet40attack success rate: 55.7 -> 79.2Improved
use CFG instead of PF-Attack against adversarial training defensePointNet source · ModelNet40attack success rate: 10.4 -> 22.6Improved
use CFG instead of PF-Attack on real-world scansPointNet source, DGCNN target · ScanObjectNNattack success rate: 40.3 -> 58.3Improved
use CFG instead of PF-Attack on state-of-the-art targetsPointNet source, CurveNet target · ModelNet40attack success rate: 22.7 -> 29.2Improved
use CFG instead of PF-Attack on DGCNN targetPointNet source · ModelNet40attack success rate: 24.8 -> 39.3Improved
use CFG instead of AdvPC when PointNet++ MSG is source and DGCNN targetPointNet++ MSG source · ModelNet40attack success rate: 39.6 -> 25.3Hurt
use CFG instead of PF-Attack (generation time)PointNet source · ModelNet40running time (min): 0.10 -> 0.07Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy