Point cloud place recognition on B.D. reference maps in-house (test)
86.49Avg Recall @ 1%PN_VLAD
Evaluation Results
| Method | Links | |
|---|---|---|
| PN_VLADArchitecture=PointNetVLAD, Training Datasets=Oxford, U.S., R.A.2018.04 | 86.49 | |
| PN_MAXArchitecture=PointNet + Max Pooling, Training Datasets=Oxford, U.S., R.A.2018.04 | 69.49 | |
| PointNetVLAD (PN_VLAD)Clusters (K)=64, Loss=lazy quadruplet loss, Output Dimension=2562018.04 | 65.3 | |
| PN_MAXLoss=lazy quadruplet loss, Output Dimension=2562018.04 | 54.74 | |
| PN_STDOutput Dimension=2562018.04 | 53.02 | |
| PN_STDArchitecture=PointNet + Standard Deviation, Training Datasets=Oxford, U.S., R.A.2018.04 | 53.02 |