AI Paper Atlas

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

Targeted Mismatch Adversarial Attack: Query With a Flower to Retrieve the Tower

Giorgos Tolias, Filip Radenovic, Ondrej Chum

ICCV 2019 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial perturbations beyond images · 72 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 Gaussian blurring before down-sampling in the attack (ablation)multi-resolution attack · unseen test resolutionsdescriptor similarity / retrievalImproved
optimize over multiple attack-resolutions instead of a single one (ablation)targeted mismatch attack · unknown test-resolutionretrieval performanceImproved
use logarithmic sampling of attack-resolutions (ablation)multi-resolution attack · unknown test-resolutionretrieval performanceMixed
increase distortion weight lambda from 0 to 10 (ablation)L_hist attack · RParisdescriptor similarity to target: 0.990 -> 0.767Hurt
increase distortion weight lambda from 0 to 10 (ablation)L_hist attack · RParisdistortion w.r.t. carrier: 0.0177 -> 0.0008Improved
use activation histogram loss instead of tensor loss (ablation)L_hist vs L_tens · optimization costiterations and distortionImproved
use activation histogram instead of activation tensor to avoid revealing the target (ablation)L_hist vs L_tens · concealmenttarget content reconstruction from tensor inversionImproved
optimize ensemble of pooling operations instead of GeM only for unknown test pooling (ablation)L_P vs L_desc(GeM) · RParis, test pooling SPoCmAP difference to original: -4.4 -> -0.1Improved
use histogram loss instead of GeM descriptor loss for unknown test pooling (ablation)L_hist vs L_desc(GeM) · RParis, test pooling SPoCmAP difference to original: -4.4 -> -0.1Improved
add descriptor whitening to the test modelL_hist dense attack · RParis, full resolutionmAP: 40.2 -> 47.3Mixed
add descriptor whitening to the test model at resolution 768L_hist dense attack · RParis, resolution 768mAP: 39.4 -> 42.9Mixed
transfer attack optimized on AlexNet+ResNet18 ensemble to unseen VGG16L_hist dense · HolidaysmAP difference to original: -72.6Hurt

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