AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking

Hongjun Wang, Guangrun Wang, Ya Li, Dongyu Zhang, Liang Lin

CVPR 2020 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training with feature consistency · 88 citations (OpenAlex)

Check 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
Apply the proposed attack to AlignedReIDAlignedReID · Market1501Rank-1: 91.8 -> 1.4Improved
Use mis-ranking loss instead of modified misclassification loss alone (ablation)AlignedReID attack · Market1501Rank-1: 13.7 -> 4.5Improved
Combine modified xent with mis-ranking loss (ablation)AlignedReID attack · Market1501Rank-1: 4.5 -> 1.4Improved
Replace conventional misclassification loss with the proposed combined losses (ablation)AlignedReID attack · Market1501Rank-1: 28.5 -> 1.4Improved
Replace PatchGAN with a multi-stage discriminator (ablation)AlignedReID attack · Market1501, ε=40Rank-1: 48.3 -> 0.0Improved
Replace PatchGAN with a multi-stage discriminator (ablation)AlignedReID attack · Market1501, ε=10Rank-1: 53.3 -> 24.4Improved
Reduce perturbation bound from ε=16 to ε=10 (ablation)AlignedReID attack · Market1501Rank-1: 1.4 -> 24.4Hurt
Constrain attack locations to pixel ratio 1/2 rather than full size (ablation)AlignedReID attack · Market1501Rank-1: 1.4 -> 39.3Hurt
Constrain attack locations to pixel ratio 1/4 without increased magnitude (ablation)AlignedReID attack · Market1501Rank-1: 1.4 -> 72.7Hurt
Increase noise magnitude at pixel ratio 1/4 (ablation)AlignedReID attack · Market1501Rank-1: 72.7 -> 0.3Improved
Use learned attack locations instead of random locations (ablation)AlignedReID attack with Gaussian noise · Market1501Rank-1: 81.9 -> 79.4Improved
Use learned attack locations instead of random locations (ablation)AlignedReID attack with uniform noise · Market1501Rank-1: 51.1 -> 50.7Improved

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