Image Retrieval On Amstertime
Metriken
mAP
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | mAP | Paper Title | Repository |
|---|---|---|---|
| DINOv2 distilled (ViT-S/14 frozen) | 43.5 | DINOv2: Learning Robust Visual Features without Supervision | |
| DINOv2 (ViT-g/14 frozen) | 46.7 | DINOv2: Learning Robust Visual Features without Supervision | |
| AP-GeM (ResNet-101) | 35 | AmsterTime: A Visual Place Recognition Benchmark Dataset for Severe Domain Shift | |
| DINOv2 distilled (ViT-B/14 frozen) | 45.6 | DINOv2: Learning Robust Visual Features without Supervision | |
| DINOv2 distilled (ViT-L/14 frozen) | 50.0 | DINOv2: Learning Robust Visual Features without Supervision |
0 of 5 row(s) selected.