AI Paper Atlas

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

Lipschitz Optimization for Formal Verification of Homographies

Jean-Guillaume Durand, Panagiotis Kouvaros, Maxime Gariel, Alessio Lomuscio

CVPR 2026 · Adversarial robustness and security › Certified and empirical robustness · approach: Neural network robustness verification · 0 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
use improved implementation (half-pixel gradients, extra Lipschitz candidates, early dichotomy, q=2, pixel parallelism) over PWLPWL · MNIST 5 degree rotationBaB steps: 124.0 -> 17.5Improved
use improved implementation over PWLPWL · MNIST 5 degree rotationtime (s): 87.6 -> 7.0Improved
use improved implementation over PWLPWL · CIFAR-10 5 degree rotationtime (s): 910.3 -> 98.7Improved
use improved implementation over PWLPWL · MNIST 20 degree rotationtime (s): 90.3 -> 36.2Improved
use improved implementation over PWL (q=2 bounds)PWL · CIFAR-10 20 degree rotationpolytope area: 15.1 -> 14.0Improved
use per-pixel parallelization at fixed BaB budgetPWL · MNIST 5 degree rotationtime (s): 87.6 -> 12.9Improved
increase yaw perturbation amplitude from 1 to 5 degreeshomography verification · MNISTverified (%): 76 -> 25Hurt
replace black padding with replicate padding (yaw [0,2.5] deg)homography verification · CIFAR-10robust cases (%): 34 -> 42Improved
replace black padding with reflect padding (yaw [0,5] deg)homography verification · CIFAR-10robust cases (%): 7 -> 9Improved
replace black padding with wrap padding (yaw [0,2.5] deg)homography verification · CIFAR-10robust cases (%): 34 -> 35No clear effect
use reflection padding instead of black paddinghomography verification · GTSRBrobust cases (%)No clear effect
evaluate roll vs yaw perturbation (affine vs projective)homography verification · CIFAR-10robust cases (%): 69 -> 7Mixed

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