Research map / Learning theory and algorithms / Prediction and decision losses
MAGSAC: Marginalizing Sample Consensus
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · kusvod2 (F) | e_avg (RMS geometric error, pixels): 0.52 -> 0.38 | Helped |
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · AdelaideRMF (F) | e_avg (RMS geometric error, pixels): 0.27 -> 0.30 | Hurt |
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · Multi-H (F) | e_avg (RMS geometric error, pixels): 0.50 -> 0.47 | Helped |
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · homogr (H) | e_avg (RMS geometric error, pixels): 1.83 -> 1.37 | Helped |
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · EVD (H) | e_avg (RMS geometric error, pixels): 3.55 -> 1.76 | Helped |
| Marginalize over noise scales and polish each candidate with σ-consensus | MAGSAC · strecha (E) | e_avg (RMS geometric error, pixels): 7.13 -> 6.51 | Helped |
| Apply σ-consensus as post-processing (ablation) | RANSAC · all datasets | e_avg (RMS geometric error, pixels): 3.07 -> 2.47 | Helped |
| Apply σ-consensus as post-processing (ablation) | MSAC · all datasets | e_avg (RMS geometric error, pixels): 3.04 -> 2.48 | Helped |
| Apply σ-consensus as post-processing (ablation) | LO-RANSAC · all datasets | e_avg (RMS geometric error, pixels): 3.30 -> 2.83 | Helped |
| Apply σ-consensus as post-processing (ablation) | LO-MSAC · all datasets | e_avg (RMS geometric error, pixels): 3.39 -> 2.88 | Helped |
Closest papers by meaning
- RANSAC Scoring Done Right (NeurIPS 2026 · similarity 0.87)
- On the Instability of Relative Pose Estimation and RANSAC's Role (CVPR 2022 · similarity 0.86)
- Learning To Find Good Models in RANSAC (CVPR 2022 · similarity 0.86)
- Latent RANSAC (CVPR 2018 · similarity 0.85)
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.83)
- The Value of Covariance Matching in Gaussian DDPMs and the Lanczos Sampler (NeurIPS 2026 · similarity 0.80)
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.