MAGSAC: Marginalizing Sample Consensus

By Daniel Barath, Jiri Matas. Computer Vision and Pattern Recognition, 2018

351 citations recorded by the source corpus.

A method called, sigma-consensus, is proposed to eliminate the need for a user-defined inlier-outlier threshold in RANSAC. Instead of estimating the noise sigma, it is marginalized over a range of noise scales. The optimized model is obtained by weighted least-squares fitting where the weights come from the marginalization over sigma of the point likelihoods of being inliers. A new quality function is proposed not requiring sigma and, thus, a set of inliers to determine the model quality. Also, a new termination criterion for RANSAC is built on the proposed marginalization approach. Applying sigma-consensus, MAGSAC is proposed with no need for a user-defined sigma and improving the accuracy of robust estimation significantly. It is superior to the state-of-the-art in terms of geometric accuracy on publicly available real-world datasets for epipolar geometry (F and E) and homography estimation. In addition, applying sigma-consensus only once as a post-processing step to the RANSAC output always improved the model quality on a wide range of vision problems without noticeable deterioration in processing time, adding a few milliseconds.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/537e1530ffdb2620135906927f55353ae5a1e021

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

MAGSAC++, a Fast, Reliable and Accurate Robust Estimator (2019, methodology): “We propose a new quality function and model fitting procedure for MAGSAC [2].”
LightGlue: Local Feature Matching at Light Speed (2023, methodology): “LightGlue thus makes DLT, a simple solver, competitive with the expensive and slower MAGSAC [3].”
Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “More recently, Barath et al. (2019b) applyed σ -consensus in their MAGSAC, to eliminate the need of a user-defined threshold by marginalizing over a range of noise scales.”
COTR: Correspondence Transformer for Matching Across Images (2021, background): “The latter are the focus of much research, whether hand-crafted, following RANSAC [18, 12, 3], consensus- or motion-based heuristics [11, 31, 6, 37], or learned [75, 7, 76, 64].”
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose (2021, methodology): “Such a formulation estimates the camera pose using a Perspective-n-Point (PnP) solver [1, 14, 31, 39, 40] inside a RANSAC loop [6, 19, 28, 43].”
Image Matching Across Wide Baselines: From Paper to Practice (2020, methodology): “On the right, we show SIFT (1999) [53] with a carefully tuned MAGSAC [15] – notice how the latter performs much better.”
A review of multimodal image matching: Methods and applications (2021, background): “Several studies address this issue in a local optimization procedure [314,315], or a progressive growing sampling procedure [311,316,317] to eliminate the need for user-defined thresholds such as inlier–outlier decision.”
ACNe: Attentive Context Normalization for Robust Permutation-Equivariant Learning (2020, methodology): “This makes our method more broadly applicable, and directly comparable with most robust estimation methods for stereo [16, 45, 9, 10, 4].”