AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

Towards Effective and Transferable Physical Camouflage against Multi-View BEV-based 3D Perception in Autonomous Driving

Linye Lyu, Jiawei Zhou, Daojing He, YU LI

NeurIPS 2026 · Detection and segmentation › 3D detection and driving · approach: Multi-camera BEV detection · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a physical camouflage attack against multi-view BEV-based 3D perception models in autonomous driving, aiming to be effective and transferable across models.

Problem

Robustness of multi-view BEV-based 3D perception models in autonomous driving against physical adversarial camouflage attacks is largely unexplored; existing camouflage attacks target single-view 2D detectors.

Approach

Designs a physical camouflage attack targeting multi-view BEV 3D perception rather than single-view 2D detectors, with the goal of effectiveness and transferability across models. Full method details are unavailable from the abstract.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply BEV mask to the BEV-feature adversarial loss (ablation)BEVCA · BEVFormer-base 3D detectionAP: 0.612 -> 0.191Improved
use BEV features instead of detection scores as attack loss (ablation)BEVCA · BEVFormer-base 3D detectionAP: 0.564 -> 0.191Improved
use masked BEV loss instead of image-encoder feature loss (ablation)BEVCA · BEVFormer-base 3D detectionAP: 0.392 -> 0.191Improved
use masked BEV loss instead of detection-head feature loss (ablation)BEVCA · BEVFormer-base 3D detectionAP: 0.471 -> 0.191Improved
unmasked BEV loss over whole BEV map (ablation)BEV-feature attack · BEVFormer-base 3D detectionAP: 0.618 normal -> 0.612Hurt
increase total variation weight beta (ablation)BEVCA · BEVFormer seriesattack performanceImproved
BEV mask thresholds delta r and delta theta set to 15 m and 15 degrees (ablation)BEVCA · BEVFormer-baseAPImproved
fine-tune victim models on BEVCA adversarial samplesBEV perception models · CARLA testAP: 0.191 -> 0.615Improved
apply JPEG compression defenseBEVFormer-base under BEVCA · CARLA testBEVCA AP: 0.191 -> 0.113 (QF=25)No clear effect
heavy Gaussian input smoothing sigma=2.0BEVFormer-base under BEVCA · CARLA testBEVCA AP: 0.191 -> 0.283Mixed
use BEVCA instead of CNCA in physical worldcamouflage texture · Physical world, 4 BEV modelsattack success rate: 30.00% -> 49.60% (BEVFormer base)Improved

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