Table 3 Registration results for scenario 1
Scenario 1 | Recall\(_C\) [%] | \(S_{Vg}\) | \(S_{Vs}\) | RMSD | |
|---|---|---|---|---|---|
Ground truth | 100 | 0.710 | 0.521 | 0.657 | |
Feature-based | SIFT + FPFH | 37.7 | 0.907 | 0.864 | 4.043 |
SIFT + PFHRGB | 25.1 | 0.945 | 0.923 | 4.449 | |
SIFT + RIFT | 14 | 0.972 | 0.966 | 4.687 | |
Deep learning | DCP | 3.8 | 0.994 | 0.993 | 4.914 |
PointNetLK | 7.4 | 0.987 | 0.983 | 4.833 | |
DeepGMR | 14.4 | 0.973 | 0.967 | 4.679 | |
GeoTransformer | 26.9 | 0.945 | 0.931 | 4.369 | |
Predator | 34 | 0.930 | 0.910 | 4.146 | |
Proposed solution | 100 | 0.711 | 0.498 | 0.680 | |