Abstract:Deep visual graph matching (GM) is a challenging combinatorial task that involves finding a permutation matrix that indicates the correspondence between keypoints from a pair of images. Like many learning systems, empirical studies have shown that visual GM is susceptible to adversarial attacks, with reliability issues in downstream applications. To the best of our knowledge, certifying robustness for deep visual GM remains an open challenge with two main difficulties: how to handle the paired inputs together with the heavily non-linear permutation output space (especially at large scale), and how to balance the trade-off between certified robustness and matching performance. Inspired by the randomized smoothing (RS) technique, we propose the Certified Robustness based on the Optimal Smoothing Range Search (CR-OSRS) technique to fulfill the robustness guarantee for deep visual GM. First, unlike conventional RS methods that use isotropic Gaussian distributions for smoothing, we build the smoothed model with paired joint Gaussian distributions, which capture the structural information among keypoints, and mitigate the performance degradation caused by smoothing. For the vast space of the permutation output, we devise a similarity-based partitioning method that can lower the computational complexity and certification difficulty. We then derive a stringent robustness guarantee that links the certified space of inputs to their corresponding fixed outputs. Second, we design a global optimization method to search for optimal joint Gaussian distributions and facilitate a larger certified space and better performance. Third, we apply data augmentation and a similarity-based regularizer in training to enhance smoothed model performance. Lastly, for the high-dimensional and multivariable nature of the certified space, we propose two methods (sampling and marginal radii) to evaluate it. Experimental results on public benchmarks show that our method achieves state-of-the-art certified robustness.

Genetic Consistency Estimation for Strong Robustness in Computer Vision

2-Entity Random Sample Consensus for Robust Visual Localization: Framework, Methods, and Verifications

Deterministic Optimality for Robust Vehicle Localization Using Visual Measurements

A highly robust estimator through partially likelihood function modeling and its application in computer vision

Model Quality Aware RANSAC: A Robust Camera Motion Estimator

Globally Optimal Consensus Maximization for Robust Visual Inertial Localization in Point and Line Map.

Robust stereo visual odometry: A comparison of random sample consensus algorithms based on three major hypothesis generators

Outlier-Robust Estimation: Hardness, Minimally Tuned Algorithms, and Applications

Robust LS-VCE for the Nonlinear Gauss–Helmert Model: Case Studies for Point Cloud Fitting and Geodetic Symmetric Transformation

Robust Kernel Recursive Adaptive Filtering Algorithms Based on M-estimate

A New Approach to Robust Fundamental Matrix Estimation Using an Analytic Objective Function and Adjusted Gradient Projection

A New Robust Approach for Remote Sensing Image Regional Classification

Robust Incremental State Estimation through Covariance Adaptation

Certified Robustness on Visual Graph Matching Via Searching Optimal Smoothing Range

RGM: A Robust Generalist Matching Model.

Research on Robustness of Geometric Algebra Based Adaptive Filtering Algorithms In Non-Gaussian Environment

Unified robust estimation

A Survey on the Robustness of Computer Vision Models against Common Corruptions

Optimal Robust Estimation under Local and Global Corruptions: Stronger Adversary and Smaller Error

RGM: A Robust Generalizable Matching Model

Robust Statistical Estimation and Segmentation of Multiple Subspaces