AI Paper Atlas

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

Indirect Local Attacks for Context-aware Semantic Segmentation Networks

Krishna Kanth Nakka, Mathieu Salzmann

ECCV 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial perturbations beyond images · 29 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
raise l_inf step size alpha from 0.00001 to 0.005 (ablation)Indirect attack on FCN · CityscapesASR_t: 5.0% -> 61%Improved
raise l_inf step size alpha from 0.00001 to 0.005 (ablation)Indirect attack on PSPNet · CityscapesASR_t: 12% -> 90%Improved
restrict perturbation to pixels at least d=150 pixels from dynamic class (vs d=0) (ablation)Indirect attack on FCN · CityscapesASR_t: 64% -> 0.0%Hurt
restrict perturbation to pixels at least d=150 pixels from dynamic class (vs d=0) (ablation)Indirect attack on PSANet · CityscapesASR_t: 90% -> 47%Hurt
restrict perturbation to pixels at least d=150 pixels from dynamic class (vs d=0) (ablation)Indirect attack on DANet · CityscapesASR_t: 90% -> 10%Hurt
use adaptive sparsity S=75% instead of S=95% perturbation mask (ablation)Adaptive indirect attack on PSANet · CityscapesASR_t: 44% -> 78%Improved
use adaptive sparsity S=75% instead of S=95% perturbation mask (ablation)Adaptive indirect attack on FCN · CityscapesASR_t: 1.0% -> 12%Improved
increase universal patch size from 1.0% to 9.0% of image (ablation)Universal patch on PSPNet · CityscapesASR_u: 3.0% -> 83%Improved
increase universal patch size from 1.0% to 9.0% of image (ablation)Universal patch on FCN · CityscapesASR_u: 2.0% -> 18%Improved
use Mahalanobis-based detector instead of SC and Re-Syn baselinesDetector on FCN, universal patch · CityscapesGlobal AUROC: 0.71 -> 1.00Improved
use Mahalanobis-based detector on global perturbations instead of Re-SynDetector on FCN, global attack · CityscapesGlobal AUROC: 1.00 -> 0.94Hurt

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