Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
BANSAC: A Dynamic BAyesian Network for Adaptive SAmple Consensus
Valter Piedade, Pedro Miraldo
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 |
|---|---|---|---|
| use 1st-order instead of higher-order Markov assumption (ablation) | BANSAC · sacre_coeur calibrated relative pose | Rotation mAA (5 deg): 0.808 -> 0.836 | Improved |
| use 3rd-order Markov assumption for speed (ablation) | BANSAC · sacre_coeur calibrated relative pose | Avg. execution time (ms): 13.9 -> 11.8 | Mixed |
| use linear sampling weights rho_1 instead of nonlinear activations (ablation) | BANSAC · sacre_coeur calibrated relative pose | Rotation mAA (5 deg): 0.823 -> 0.836 | Improved |
| choose CPT activation gamma_1 over gamma_2/gamma_3 (ablation) | BANSAC · sacre_coeur calibrated relative pose | Rotation mAA (5 deg): 0.790 -> 0.836 | Improved |
| combine BANSAC stopping criterion with SPRT or RANSAC (ablation) | BANSAC · sacre_coeur calibrated relative pose | Avg. execution time (ms): 33.5 -> 13.9 | Improved |
| replace RANSAC sampling with BANSAC dynamic Bayesian updates on calibrated pose | RANSAC · PhotoTourism calibrated relative pose | Avg. execution time (ms): 25.5 -> 15.6 | Improved |
| use BANSAC on homography estimation versus RANSAC runtime | RANSAC · Homography estimation | Avg. execution time (ms): 2.30 -> 4.04 | Hurt |
| add local optimization to all methods | RANSAC family · Uncalibrated relative pose | Rotation mAA (5 deg): 0.467 -> 0.514 (RANSAC) | Improved |
| BANSAC under decreasing inlier ratio | BANSAC · Uncalibrated relative pose, varying inlier ratio | accuracy | Improved |
Closest papers by meaning
- RANSAC Scoring Done Right (NeurIPS 2026 · similarity 0.90)
- Marginalizing Sample Consensus (TPAMI 2021 · similarity 0.89)
- MAGSAC: Marginalizing Sample Consensus (CVPR 2019 · similarity 0.89)
- Latent RANSAC (CVPR 2018 · similarity 0.89)
- Monte Carlo Diffusion for Generalizable Learning-Based RANSAC (AAAI 2026 · similarity 0.89)
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.86)
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.
