AI Paper Atlas

Research map / Point clouds / Point cloud classification

Fooling LiDAR Perception via Adversarial Trajectory Perturbation

Yiming Li, Congcong Wen, Felix Juefei-Xu, Chen Feng

ICCV 2021 · Point clouds › Point cloud classification · approach: Adversarial point cloud attacks · 74 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
attack classification of stage-2 instead of random translation noiseFLAT (Translation) · nuScenes car EasyAP: additional drops of 5.28, 2.71, and 2.84 in AP versus random perturbation (easy, moderate, hard)Improved
perturb rotation instead of translationFLAT · nuScenes car EasyAP: 11.72 -> 2.35Improved
perturb full trajectory instead of rotation onlyFLAT · nuScenes car EasyAP: 2.35 -> 0.19Improved
attack stage-2 classification versus stage-2 regression (translation)FLAT (Translation) · nuScenes car EasyAP: 26.09 -> 11.72Improved
attack with polynomial trajectory perturbation instead of per-point translationFLAT · nuScenes car EasyAP (stage-2 cls): 11.72 -> 12.51No clear effect
attack with coordinate-level perturbation instead of trajectory translationPointRCNN attack · nuScenes car EasyAP: 11.72 -> 16.42Hurt
add trajectory smoothness regularization (lambda 0.01) (ablation)FLAT · nuScenes car EasyAP: 0.19 -> 0.28No clear effect
add point cloud l_p regularization lambda=0.01 (ablation)FLAT · nuScenes car EasyAP: 0.19 -> 5.66Mixed
increase point cloud regularization strength from 0.01 to 1 (ablation)FLAT · nuScenes car EasyAP: 5.66 -> 23.65Hurt
transfer attack to black-box PointPillar++ with full trajectory perturbationFLAT · nuScenes valmAP: 35.4 -> 14.8Improved
attack rotation instead of translation in black-box transferFLAT · nuScenes valmAP: 20.6 -> 16.4Improved

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