Visual Place Recognition On Nardo Air R
Metriken
Recall@1
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| AnyLoc-VLAD-DINO | 94.37 | AnyLoc: Towards Universal Visual Place Recognition |
| CosPlace | 91.55 | Rethinking Visual Geo-localization for Large-Scale Applications |
| AnyLoc-VLAD-DINOv2 | 85.92 | AnyLoc: Towards Universal Visual Place Recognition |
| DINO | 84.51 | Emerging Properties in Self-Supervised Vision Transformers |
| MixVPR | 76.06 | MixVPR: Feature Mixing for Visual Place Recognition |
| DINOv2 | 71.83 | DINOv2: Learning Robust Visual Features without Supervision |
| CLIP | 61.97 | AnyLoc: Towards Universal Visual Place Recognition |
| NetVLAD | 60.56 | NetVLAD: CNN architecture for weakly supervised place recognition |
0 of 8 row(s) selected.