Table 3 Unsupervised pose estimation results with different detection models on large pose dataset.
From: Robust pose estimation for non-cooperative space objects based on multichannel matching method
Methods | mAUC | APE | mMS |
|---|---|---|---|
Sift-Superglue | 0.355 | 7.391 | 0.115 |
HardNet-smnn | 0.334 | 13.18 | 0.000 |
KeyNet-mnn | 0.278 | 10.08 | 0.090 |
LoFTR | 0.415 | 4.354 | 0.165 |
Hardnet-Superglue | 0.420 | 4.985 | 0.135 |
Sift-mnn | 0.136 | 17.36 | 0.044 |
Superpoint-Superglue | 0.208 | 30.02 | 0.069 |
Proposed | 0.636 | 1.776 | 0.418 |
