AI Paper Atlas

Research map / 3D vision / Gaussian splatting

3D Gaussian Splatting Driven Multi-View Robust Physical Adversarial Camouflage Generation

Tianrui Lou, Xiaojun Jia, Siyuan Liang, Jiawei Liang, Ming Zhang, Yanjun Xiao, Xiaochun Cao

ICCV 2025 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 6 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 multi-view camouflage consistency (ablation)PGA · CARLA digital ablationAP@0.5 average (lower is better): 27.32 -> 22.87Improved
add min-max optimization with background counter-attack (ablation)PGA · CARLA digital ablationAP@0.5 average (lower is better): 30.50 -> 22.87Improved
add multi-view consistency (ablation)PGA · CARLA digital, Faster R-CNNAP@0.5: 8.05 -> 3.57Improved
add min-max optimization (ablation)PGA · CARLA digital, Faster R-CNNAP@0.5: 10.23 -> 3.57Improved
PGA instead of RAUCA at 5mPGA · CARLA digital, 5mAP@0.5 average: 37.16 -> 23.46Improved
PGA instead of RAUCA at 10mPGA · CARLA digital, 10mAP@0.5 average: 39.25 -> 21.78Improved
PGA instead of RAUCA at 20mPGA · CARLA digital, 20mAP@0.5 average: 49.50 -> 20.73Improved
PGA instead of RAUCA across pitch anglesPGA · CARLA pitch sweepAP@0.5 average: 33.97 -> 6.73Improved
PGA instead of RAUCA on 1:24 toy car at 50cmPGA · 1:24 physical, 50cmAP@0.5 average: 36.69 -> 28.69Improved
PGA instead of RAUCA on 1:24 toy car at 100cmPGA · 1:24 physical, 100cmAP@0.5 average: 37.25 -> 28.26Improved
deploy PGA camouflage on real vehiclePGA · 1:1 real vehicleAP@0.5 Faster R-CNN: 88.48 -> 25.67Improved
transfer PGA from Faster R-CNN to YOLO-v5 (5m sunny)PGA · CARLA digital, 5mAP@0.5: 4.52 -> 39.10Mixed

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