Random sample consensus (Fischler and Bolles, 1981): fit a model to random minimal subsets of the data and keep the one most points agree with. The workhorse of outlier rejection in registration and loop closure, and the baseline that certifiable methods such as TEASER++ are measured against.
Mentioned in
10 papers:
- vS-Graphs (2026)
- Picasso (2026)
- DRACo-SLAM2 (2025)
- Hydra-Multi (2023)
- S-Graphs (2022)
- Hydra (2022)
- TEASER++ (2021)
- GNC (2020)
- 3D Dynamic Scene Graphs (2020)
- PCM (2018)