AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Evaluating the Adversarial Robustness of Semantic Segmentation: Trying Harder Pays Off

Levente Halmosi, Bálint Mohos, Márk Jelasity

ECCV 2024 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision · 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
evaluate with an aggregated worst-case attack over ten attacks rather than a single attackpublished robust segmentation models (DDC-AT, PGD-AT, SegPGD-AT) · Cityscapes and PASCAL VOCrobust accuracy / mIoU: robust values near 0.00 for 50% adversarial batch modelsHurt
train with 100% adversarial batches instead of 50% (ablation)PGD-AT / SegPGD-AT · Cityscapes DeepLabv3robust NmIoU: 0.01 -> 15.86 (PGD-AT); 0.00 -> 21.82 (SegPGD-AT)Improved
train with 100% adversarial batches instead of 50% (ablation)PGD-AT / SegPGD-AT · Cityscapes DeepLabv3clean NmIoU: 51.18 -> 32.68 (Normal vs SegPGD-AT-100)Hurt
compute mIoU as image-wise average (NmIoU) instead of class-wise global IoU (CmIoU) (ablation)SEA-AT models · PASCAL VOCrobust mIoU: 32.79 -> 13.20 (SEA-AT-Tiny CmIoU vs NmIoU)Hurt
use PAdam-Cos as an individual attack on normally trained modelsNormal models · Cityscapes DeepLabv3robust NmIoU: 0.00Improved
run the aggregated ten-attack ensemble versus the best single attack (ablation)SegPGD-AT-100 · Cityscapes DeepLabv3robust NmIoU: 23.11 -> 21.82Improved
use SEA attacks rather than PAdam for robust models (ablation)robust segmentation models · Cityscapes DeepLabv3robust NmIoUMixed
remove background pixels from the metric (ablation)PGD-AT-100 and SegPGD-AT-100 · PASCAL VOCforeground robustnessHurt

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