mlsanityData to 14 Oct 2026

MAGSAC: marginalizing sample consensus

arXiv 1803.07469 ↗First posted 20 Mar 2018cs.CVcited by 123

Not on the map. Why?

Abstract (arXiv)

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. The source code is at https://github.com/danini/magsac.

Key facts. Cited by 123 papers in the map (data through 14 Oct 2026). First posted on arXiv 20 Mar 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 18; 2022: 15; 2023: 24; 2024: 26; 2025: 21; 2026: 19.

123 papers.

Sort:

By area:

  1. 2023LightGlue: Local Feature Matching at Light Speedcited by 590
  2. 2021Back to the Feature: Learning Robust Camera Localization from Pixels to Posecited by 199
  3. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 183
  4. 2022Map-free Visual Relocalization: Metric Pose Relative to a Single Imagecited by 114
  5. 2024Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localizationcited by 82
  6. 2022OSOP: A Multi-Stage One Shot Object Pose Estimation Frameworkcited by 81
  7. 2021PDC-Net+: Enhanced Probabilistic Dense Correspondence Networkcited by 80
  8. 2024GIM: Learning Generalizable Image Matcher From Internet Videoscited by 75
  9. 2022SC^2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registrationcited by 62
  10. 2024PF3plat: Pose-Free Feed-Forward 3D Gaussian Splattingcited by 62
  11. 2023SiLK -- Simple Learned Keypointscited by 54
  12. 2024GLACE: Global Local Accelerated Coordinate Encodingcited by 48
38 more
  1. 2021Progressive Correspondence Pruning by Consensus Learningcited by 46
  2. 2022Iterative Deep Homography Estimationcited by 45
  3. 2023Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairscited by 37
  4. 2022Generalized Differentiable RANSACcited by 30
  5. 2021Motion Basis Learning for Unsupervised Deep Homography Estimation with Subspace Projectioncited by 28
  6. 2021LocalTrans: A Multiscale Local Transformer Network for Cross-Resolution Homography Estimationcited by 28
  7. 2021Depth-Aware Multi-Grid Deep Homography Estimation with Contextual Correlationcited by 28
  8. 2025Emergent Outlier View Rejection in Visual Geometry Grounded Transformerscited by 22
  9. 2023Visual Localization using Imperfect 3D Models from the Internetcited by 20
  10. 2022End2End Multi-View Feature Matching with Differentiable Pose Optimizationcited by 19
  11. 2022Unsupervised Homography Estimation with Coplanarity-Aware GANcited by 19
  12. 2023DeepMatcher: A Deep Transformer-based Network for Robust and Accurate Local Feature Matchingcited by 18
  13. 2024MADiff: Motion-Aware Mamba Diffusion Models for Hand Trajectory Prediction on Egocentric Videoscited by 18
  14. 2021Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimationcited by 17
  15. 2021VSAC: Efficient and Accurate Estimator for H and Fcited by 17
  16. 2024Scalable 3D Registration via Truncated Entry-wise Absolute Residualscited by 16
  17. 2022ARCS: Accurate Rotation and Correspondence Searchcited by 15
  18. 2025Convex Relaxation for Robust Vanishing Point Estimation in Manhattan Worldcited by 15
  19. 2023Long-term Visual Localization with Mobile Sensorscited by 13
  20. 2024LoD-Loc: Aerial Visual Localization using LoD 3D Map with Neural Wireframe Alignmentcited by 13
  21. 2021ScaleNet: A Shallow Architecture for Scale Estimationcited by 12
  22. 2024FAR: Flexible, Accurate and Robust 6DoF Relative Camera Pose Estimationcited by 12
  23. 2025OrthoLoC: UAV 6-DoF Localization and Calibration Using Orthographic Geodatacited by 12
  24. 2023Two-View Geometry Scoring Without Correspondencescited by 11
  25. 2026LoMa: Local Feature Matching Revisitedcited by 11
  26. 2022Semi-supervised Deep Large-baseline Homography Estimation with Progressive Equivalence Constraintcited by 10
  27. 2023GyroFlow+: Gyroscope-Guided Unsupervised Deep Homography and Optical Flow Learningcited by 10
  28. 2023IMP: Iterative Matching and Pose Estimation with Adaptive Poolingcited by 10
  29. 2024MGNet: Learning Correspondences via Multiple Graphscited by 10
  30. 2022NeFSAC: Neurally Filtered Minimal Samplescited by 9
  31. 2023Learning Accurate Template Matching with Differentiable Coarse-to-Fine Correspondence Refinementcited by 9
  32. 2023Vanishing Point Estimation in Uncalibrated Images with Prior Gravity Directioncited by 9
  33. 2023VSFormer: Visual-Spatial Fusion Transformer for Correspondence Pruningcited by 8
  34. 2024Efficient and Robust Point Cloud Registration via Heuristics-guided Parameter Searchcited by 8
  35. 2024A Subspace-Constrained Tyler's Estimator and its Applications to Structure from Motioncited by 8
  36. 2024SCPNet: Unsupervised Cross-modal Homography Estimation via Intra-modal Self-supervised Learningcited by 7
  37. 2021DeepOIS: Gyroscope-Guided Deep Optical Image Stabilizer Compensationcited by 6
  38. 2021Space-Partitioning RANSACcited by 6

50 most cited of 123 All 123 (JSON)