Point Cloud Registration On Fpv1
Metrics
RRE (degrees)
RTE (cm)
Recall (3cm, 10 degrees)
Results
Performance results of various models on this benchmark
Model Name | RRE (degrees) | RTE (cm) | Recall (3cm, 10 degrees) | Paper Title | Repository |
|---|---|---|---|---|---|
| FCGF + YOHO-C | 3.653 | 1.668 | 29.18 | You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors | |
| DIP | 4.058 | 2.052 | 54.81 | Distinctive 3D local deep descriptors | |
| SpinNet | 3.105 | 1.670 | 42.46 | SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration | |
| GeoTransformer | 2.423 | 1.581 | 56.15 | Geometric Transformer for Fast and Robust Point Cloud Registration | |
| FCGF + PointDSC | 3.354 | 1.793 | 47.85 | PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency | |
| Greedy Grid Search | 0.014 | 0.009 | 92.81 | Challenging the Universal Representation of Deep Models for 3D Point Cloud Registration | |
| FPFH-8M | 4.347 | 1.900 | 9.51 | Fast Point Feature Histograms (FPFH) for 3D Registration | |
| FCGF + YOHO-O | 4.489 | 1.852 | 18.91 | You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors |
0 of 8 row(s) selected.