3D Place Recognition On Cs Campus3D
Metrics
AR@1
AR@1 cross-source
AR@1%
AR@1% cross-source
Results
Performance results of various models on this benchmark
Model Name | AR@1 | AR@1 cross-source | AR@1% | AR@1% cross-source | Paper Title | Repository |
|---|---|---|---|---|---|---|
| CrossLoc3D | 74.12 | 70.73 | 83.04 | 85.74 | CrossLoc3D: Aerial-Ground Cross-Source 3D Place Recognition | |
| Minkloc3D-S | 44.39 | 32.17 | 59.33 | 71.88 | MinkLoc3D-SI: 3D LiDAR place recognition with sparse convolutions, spherical coordinates, and intensity | |
| Minkloc3Dv2 | 67.06 | 52.46 | 76.68 | 83.48 | MinkLoc++: Lidar and Monocular Image Fusion for Place Recognition | |
| LPD-Net | 45.94 | 11.99 | 59.49 | 40.70 | LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis | |
| Minkloc3D | 69.38 | 55.36 | 79.1 | 85.22 | MinkLoc3D: Point Cloud Based Large-Scale Place Recognition | |
| transloc3d | 58.16 | 42.97 | 69.04 | 80.64 | TransLoc3D : Point Cloud based Large-scale Place Recognition using Adaptive Receptive Fields | |
| PointNetVLAD | 35.57 | 19.07 | 41.46 | 43.53 | PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition |
0 of 7 row(s) selected.