Table 6 Registration results for scenario 4
Scenario 4 | Recall\(_C\) [%] | \(S_{Vg}\) | \(S_{Vs}\) | RMSD | |
|---|---|---|---|---|---|
Ground truth | 100 | 0.650 | 0.551 | 0.884 | |
Feature-based | SIFT + FPFH | 11.8 | 0.980 | 0.976 | 4.767 |
SIFT + PFHRGB | 10.4 | 0.982 | 0.977 | 4.790 | |
SIFT + RIFT | 12 | 0.980 | 0.975 | 4.761 | |
Deep learning | DCP | 3.3 | 0.995 | 0.994 | 4.937 |
PointNetLK | 10.8 | 0.982 | 0.977 | 4.781 | |
DeepGMR | 13.4 | 0.978 | 0.970 | 4.728 | |
GeoTransformer | 33.3 | 0.941 | 0.916 | 4.293 | |
Predator | 29.3 | 0.949 | 0.933 | 4.384 | |
Proposed solution | 84.4 | 0.734 | 0.644 | 1.849 | |