Random sample consensus (RANSAC)
RANSAC is an iterative method to estimate parameters of a mathematical model from a set of observed data that contains outliers.
RANSAC is good for detecting outliers.
Resources
- https://en.wikipedia.org/wiki/Random_sample_consensus
- Found from the MIT course https://ocw.mit.edu/courses/16-485-visual-navigation-for-autonomous-vehicles-vnav-fall-2020/pages/lecture-notes/
- RANSAC - Random Sample Consensus (Cyrill Stachniss)
Rahul also mentioned this in the F1TENTH lecture on Scan Matching for outlier rejection.
Ransac Algorithm
Steps
- Sample the number of data points required to fit the model
- Compute model parameters using the sampled data points
- Score by the fraction of inliers within a preset threshold of the model
Repeat 1-3 until the best model is found with high confidence.
Taken from the Cyrill Stachniss video.