AI Paper Atlas

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

Stereopagnosia: Fooling Stereo Networks with Adversarial Perturbations

Alex Wong, Mukund Mundhra, Stefano Soatto

AAAI 2021 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision · 15 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 optimized FGSM perturbations with ε = 0.02 to both stereo imagesPSMNet · KITTI 2015 valD1-all error: 3.27% -> 97.33%Hurt
Use I-FGSM instead of one-step FGSM for the attackPSMNet · KITTI 2015 val, ε = 0.02D1-all error: 97.33% -> 100%Hurt
Use iterative attacks rather than random Gaussian or uniform noisePSMNet, DeepPruner, and AANet · KITTI 2015 valD1-all errorHurt
Add input diversity through random resizing and padding when crafting iterative perturbations (ablation)I-FGSM and MI-FGSM transfer attacks · KITTI 2015 val, cross-model transferD1-all error on target modelsMixed
Replace FGSM with I-FGSM for cross-model transfer (ablation)Perturbations transferred to PSMNet · KITTI 2015 valD1-all error: 55.47% -> 31.20%; 53.55% -> 48.08%Hurt
Fine-tune with clean images mixed with FGSM examples at ε = 0.02 (ablation)PSMNet · KITTI 2015 val, attack used in trainingD1-all error: 97.33% -> 2.74%Improved
Fine-tune with clean images mixed with smaller FGSM examples at ε = 0.002 (ablation)PSMNet · KITTI 2015 val, new FGSM attack ε = 0.02D1-all error: 23.74%Improved
Fine-tune on large-norm adversarial examples rather than small-norm examples (ablation)PSMNet · KITTI 2015 val, new FGSM attack ε = 0.02D1-all error: 13.84% vs 23.74%Hurt
Fine-tune with clean and small-norm adversarial examples (ablation)PSMNet and AANet · KITTI 2015 val, clean imagesD1-all errorNo clear effect
Fine-tune with large-norm adversarial examples instead of small-norm examples (ablation)Stereo models · KITTI 2015 val, clean imagesD1-all error: ≈0.4%Hurt
Apply Gaussian blur to adversarial or clean stereo images (ablation)Stereo networks · KITTI 2015 valD1-all errorHurt
Perturb both stereo images rather than only one (ablation)PSMNet, DeepPruner, and AANet · KITTI 2015 valD1-all errorImproved

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