Research map / Learning theory and algorithms / Prediction and decision losses
Marginalizing Sample Consensus
Dániel Baráth, Jana Nosková, Jiří Matas
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add sigma-consensus post-processing | RANSAC · EVD (H, 15 pairs) | e_avg (pixels): 5.73 -> 4.08 | Improved |
| add sigma-consensus post-processing | LO-MSAC · homogr (H, 16 pairs) | e_avg (pixels): 3.53 -> 2.19 | Improved |
| add sigma-consensus post-processing | RANSAC · kusvod2 (F, 24 pairs) | e_avg (pixels): 0.73 -> 0.60 | Improved |
| add sigma-consensus post-processing | LO-MSAC · AdelaideRMF (F, 19 pairs) | e_avg (pixels): 0.31 -> 0.31 | No clear effect |
| add sigma-consensus post-processing | RANSAC · All 545 image pairs | e_avg (pixels): 3.07 -> 2.47 | Improved |
| add sigma-consensus post-processing | RANSAC · All 545 image pairs | e_med (pixels): 2.17 -> 1.36 | Improved |
| add sigma-consensus post-processing | RANSAC · All 545 image pairs | failure rate: 0.19 -> 0.16 | Improved |
| use MAGSAC instead of LO-MSAC | LO-MSAC · strecha (E, 467 pairs) | e_avg (pixels): 10.62 -> 6.51 | Improved |
| use MAGSAC instead of RANSAC | RANSAC · EVD (H, 15 pairs) | e_avg (pixels): 5.73 -> 1.76 | Improved |
| use MAGSAC instead of LO-RANSAC + sigma | LO-RANSAC · AdelaideRMF (F, 19 pairs) | e_avg (pixels): 0.27 -> 0.30 | Hurt |
| use MAGSAC instead of LO-MSAC | LO-MSAC · All 545 image pairs | processing time (ms): 578 -> 688 | Hurt |
| use MAGSAC instead of RANSAC at high noise | RANSAC · synthetic homography fitting | processing time | Mixed |
Closest papers by meaning
- MAGSAC: Marginalizing Sample Consensus (CVPR 2019 · similarity 0.98)
- BANSAC: A Dynamic BAyesian Network for Adaptive SAmple Consensus (ICCV 2023 · similarity 0.89)
- RANSAC Scoring Done Right (NeurIPS 2026 · similarity 0.89)
- Learning To Find Good Models in RANSAC (CVPR 2022 · similarity 0.86)
- Monte Carlo Diffusion for Generalizable Learning-Based RANSAC (AAAI 2026 · similarity 0.86)
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.82)
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.
