Voting the outliers away

A point set with a hidden true line and adjustable outliers. One button draws a single random candidate and counts its inliers; the other runs the adaptive loop until the N(w) formula says 99% confidence has been reached.

Try this

Click on the canvas to add outliers and watch the required iteration count climb — slowly at first, then not slowly at all.

Try it: vote away the outliers

Where this lab comes from

RANSAC: How Vision Algorithms Vote Away Outliers

RANSAC separates data into inliers and outliers by fitting models to random minimal samples and keeping the one with the largest consensus. Walked end to end on two vision problems: a walking-speed estimate that three false detections drag from 2.00 m/s to 0.78, and a basketball shot called from a 3-point parabola fit. Plus the one-line formula for how many samples you need, a hands-on lab, and real inlier ratios measured on the temple dataset.

More in cameras and geometry