Improving Multisensor Positioning of Land Vehicles with Integrated Visual Odometry for Next-Generation Self-Driving Cars

Figure 3

Comparison of feature inliers using different algorithms for outlier rejection. 2399 points were detected overall. MLESAC in (b) had 1477 inliers and took 0.106 s. R-AVO-G in (c) had 2177 inliers and took 0.044 s. R-AVO-S in (d) had 1268 inliers and took 0.0008 s.

(a) All detected features

(b) MLESAC outlier rejection

(c) R-AVO-G outlier rejection

(d) R-AVO-S outlier rejection