AI Paper Atlas

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

3D Adversarial Augmentations for Robust Out-of-Domain Predictions

Alexander Lehner, Stefano Gasperini, Alvaro Marcos-Ramiro, Michael Schmidt, Nassir Navab, Benjamin Busam, Federico Tombari

IJCV 2023 · Detection and segmentation › 3D detection and driving · approach: 3D objectness and scene segmentation · 5 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
add untargeted car adversarial augmentationCylinder3D · SemanticKITTI to WaymomIoU: 29.43 -> 37.74Improved
add untargeted car adversarial augmentationCylinder3D · SemanticKITTI to nuScenesmIoU: 29.97 -> 31.55Improved
add untargeted car adversarial augmentationCylinder3D · SemanticKITTI val (in-domain)mIoU: 64.39 -> 63.85Hurt
use sample-specific iterative gradient L2 attack as augmentationCylinder3D · SemanticKITTI to WaymomIoU: 29.43 -> 24.86Hurt
use sample-specific Chamfer attack as augmentationCylinder3D · SemanticKITTI to nuScenesmIoU: 29.97 -> 22.40Hurt
add targeted (car to bicycle) augmentationCylinder3D · SemanticKITTI val bicycleIoU: 47.15 -> 54.27Improved
add untargeted person adversarial augmentationCylinder3D · SemanticKITTI to Waymo personIoU: 6.69 -> 22.22Improved
add adversarial augmentation (ours) to PointPillarsPointPillars · KITTI to WaymoAP: 40.86 -> 44.61Improved
add adversarial augmentation (ours) to Part-A2Part-A2 · KITTI to WaymoAP: 49.76 -> 56.08Improved
add our augmentation on top of SN domain adaptationPointPillars + SN · KITTI to WaymoAP: 49.27 -> 51.32Improved
add adversarial augmentation (ours) on rare crash carsPointPillars · KITTI to CrashD rare crashAP: 22.48 -> 30.37Improved
add adversarial removal augmentationPointPillars · KITTI to WaymoAP: 40.86 -> 40.32No clear effect

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