Explainability-aware frustum attack: exposing structural vulnerabilities in lidar-based 3D object detectors
File(s) Main_paper.pdf (5.11 MB)
Accepted version
Author(s)
You, Charles
Xu, binbin
Demetriou, soteris
Type
Conference Paper
Abstract
The structural vulnerabilities of point cloud–based 3D object detectors remain poorly understood. Prior work has studied adversarial robustness primarily on isolated 3D object models, while recent LiDAR spoofing attacks target richer and more realistic driving scenes but focus mainly on physical realizability rather than understanding detector be havior or attack efficiency. In this work, we investigate how LiDAR-based detectors rely on spatial evidence in complex scenes and whether these reliance patterns can be exploited to induce failures more efficiently. To this end, we propose an explainability-guided adversarial analysis methodology. We introduce the Saliency-LiDAR (SALL) method, which aggregates In tegrated Gradient attributions across scenes to produce universal saliency maps for LiDAR-based 3D object detectors. Guided by these maps, we design the Explainability-aware Frustum Attack (EFA), which selectively perturbs only the most influential frustums rather than uniformly attacking entire object regions. Experiments on KITTI and nuScenes, across detectors such as PointPillars and SECOND, show that EFA reduces detection recall by more than 15 percentage points while requiring 25–50% fewer perturbed frustums than the SOTA non–saliency-aware baseline. These findings reveal that modern 3D detectors concentrate discriminative evidence in a small subset of spatial regions, exposing a structural robustness vulnerability in current LiDAR perception systems. Our code is released at https://github.com/SecMindLab/Saliency_LiDAR.
Date Issued
2026-09-10
Date Acceptance
2026-06-18
Citation
Computer Vision – ECCV 2026, 2026, 17026, pp.398-415
ISBN
978-3-032-37594-0
Publisher
Springer
Start Page
398
End Page
415
Journal / Book Title
Computer Vision – ECCV 2026
Volume
17026
Copyright Statement
Copyright © 2026 The Author(s), under exclusive license to Springer Nature Switzerland AG. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
License URL
Identifier
http://in/chengzeng-you-75640819a
Source
European Conference on Computer Vision (ECCV)
Place of Publication
Malmö, Sweden
Publication Status
Published
Start Date
2026-09-08
Finish Date
2026-09-12
Coverage Spatial
Malmos, Sweden
Date Publish Online
2026-09-10
