AI Paper Atlas

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

Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspective

Seong Joon Oh, Mario Fritz, Bernt Schiele

ICCV 2017 · Adversarial robustness and security › Attacks on text and graphs · approach: Diffusion-based adversarial purification · 150 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 vaccination against blur (/B) to GAMAN (ablation)GAMAN · PIPA val1 GoogleNet, recogniser Brecognition rate (%): 22.2 -> 5.8Improved
add noise vaccination (/N) to GAMAN (ablation)GAMAN · PIPA val1 GoogleNet, recogniser Nrecognition rate (%): 15.0 -> 4.6Improved
add noise vaccination (/N) to GAMAN (ablation)GAMAN · PIPA val1 GoogleNet, recogniser Brecognition rate (%): 22.2 -> 23.6Hurt
use combined vaccination /TNBC instead of /B (ablation)GAMAN · PIPA val1 GoogleNet, recogniser Brecognition rate (%): 5.8 -> 9.5Hurt
use gradient vector instead of gradient signFGV vs FGS · PIPA val1 AlexNetrecognition rate (%): 23.6 -> 13.3Improved
use gradient vector instead of gradient sign (multi-iteration)GA vs BI · PIPA val1 AlexNetrecognition rate (%): 1.2 -> 0.2Improved
use multi-iteration instead of single-iteration gradientGA vs FGV · PIPA val1 AlexNetrecognition rate (%): 13.3 -> 0.2Improved
continue past decision boundary with fixed steps (GAMAN) instead of early stopping (DF)GAMAN vs DF · PIPA val1 GoogleNet, recogniser Procrecognition rate (%): 51.8 -> 4.0Improved
add basic resize and quantisation processing against GAGA · PIPA val1 GoogleNetrecognition rate (%): 0.0 -> 8.6Hurt
crop and resize applied against GAMANGAMAN · PIPA val1 ResNet152recognition rate (%): 0.0 -> 31.8Hurt
play randomised mixed strategy instead of deterministic /Buser strategy · PIPA val1 GoogleNet gamerecognition rate (%): 8.6 -> 7.3Improved
user lacks knowledge of recogniser strategy N/B user strategy · PIPA val1 GoogleNet gamerecognition rate (%): 5.8 -> 8.6Hurt

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