AI Paper Atlas

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

ODDR: Outlier Detection & Dimension Reduction Based Defense Against Adversarial Patches

Nandish Chattopadhyay, Amira Guesmi, Muhammad Abdullah Hanif, Bassem Ouni, Muhammad Shafique

ICCV 2025 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision · 8 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
ODDR defense added against GoogleAPResNet-152 · ImageNetrobust accuracy (%): 39.3 -> 79.1Improved
ODDR added against LAVANResNet-152 · ImageNetrobust accuracy (%): 10.1 -> 74.1Improved
ODDR added against DorPatchResNet-50 · ImageNetrobust accuracy (%): 4.5 -> 59.8Improved
ODDR added against SSAPMIMdepth · KITTImean depth estimation error: 0.59 -> 0.11Improved
ODDR instead of LGSResNet-152 GAP · ImageNetrobust accuracy (%): 53.9 -> 79.1Improved
ODDR instead of Dimension Reduction only (no outlier detection) (ablation)ResNet-152 GAP · ImageNetrobust accuracy (%): 66.2 -> 79.1Improved
Isolation Forest instead of DBSCAN for outlier detection (ablation)ODDR · ImageNetoverlap of detected cluster with patch: up to 20%Improved
Dimension Reduction instead of naive pixel replacement (lowest, highest, mean) (ablation)ODDR neutralization · ImageNetaccuracy: up to 15%Improved
Adaptive distribution-mapping attack against ODDRODDR ResNet-50 · ImageNetaccuracy (%): clean 78 -> 73, robust 76Mixed
Zeroth-order gradient adaptive attack, delta 0.001-0.2ODDR · ImageNetadversarial accuracy: within 5-10% of clean accuracyNo clear effect
Low-dimensional patch (99% variability preserved) adaptive attackODDR · ImageNetattack effectiveness: 99% variabilityHurt
ODDR instead of Jedi on naturalistic patchYOLO · CASIArobust average precision: 67% -> 87.5%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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