Multi-instance 3D registration on Synthetic data
53.39MHREfficient Correspondence Clustering
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Efficient Correspondence ClusteringFeature Descriptor=PREDATOR, Estimated outlier ratio=94.32%2021.11 | 53.39 | 61.44 | 51.8 | 1.28 | |
| Efficient Correspondence ClusteringFeature Descriptor=D3Feat, Estimated outlier ratio=99.30%2021.11 | 16.98 | 27.05 | 17.91 | 0.68 | |
| Progressive-XFeature Descriptor=PREDATOR, Estimated outlier ratio=94.32%2021.11 | 15.9 | 31.01 | 18.98 | 86.39 | |
| Progressive-XFeature Descriptor=D3Feat, Estimated outlier ratio=99.30%2021.11 | 4.29 | 15.28 | 5.94 | 87.22 | |
| T-LinkageFeature Descriptor=PREDATOR, Estimated outlier ratio=94.32%2021.11 | 0.19 | 0.54 | 0.27 | 43.46 | |
| CONSACFeature Descriptor=D3Feat, Estimated outlier ratio=99.30%2021.11 | 0.13 | 0.04 | 0.05 | 9.53 | |
| CONSACFeature Descriptor=PREDATOR, Estimated outlier ratio=94.32%2021.11 | 0.1 | 0.07 | 0.08 | 7.65 | |
| T-LinkageFeature Descriptor=D3Feat, Estimated outlier ratio=99.30%2021.11 | 0.07 | 0.29 | 0.1 | 56.37 |