Visual Place Recognition on St Lucia
100R@1BoQ
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| BoQBackbone=ResNet-502024.05 | 100 | — | — | |
| SALAD2024.05 | 100 | 100 | 100 | |
| BoQ2024.05 | 100 | 100 | 100 | |
| EffoVPR2024.05 | 100 | 100 | 100 | |
| SALAD2025.04 | 100 | 100 | 100 | |
| CricaVPR2025.04 | 99.9 | 99.9 | 99.9 | |
| FoLmode=global2025.04 | 99.9 | 100 | 100 | |
| FoLmode=re-ranking2025.04 | 99.9 | 100 | 100 | |
| SelaVPRreranking=true2024.02 | 99.8 | 100 | 100 | |
| SelaVPR2025.04 | 99.8 | 100 | 100 | |
| Conv-APBackbone=ResNet-50, Desc. Dim.=81922023.08 | 99.7 | — | — | |
| R2 FormerDescriptor dimension=256, Train set=MSLS2024.06 | 99.7 | — | — | |
| ProGeo (CNN)Descriptor dimension=512, Train set=SF-XL2024.06 | 99.7 | 99.9 | — | |
| Conv-APBackbone=ResNet-502024.05 | 99.7 | — | — | |
| MixVPRLearning Protocol=Supervised Learning2026.01 | 99.7 | 100 | — | |
| CosPlaceBackbone=ResNet-50, Desc. Dim.=20482023.08 | 99.6 | — | — | |
| EigenPlacesBackbone=ResNet-50, Desc. Dim.=20482023.08 | 99.6 | — | — | |
| Conv-APBackbone=ResNet-50, Desc. Dim.=40962023.08 | 99.6 | — | — | |
| MixVPRBackbone=ResNet-50, Desc. Dim.=40962023.08 | 99.6 | — | — | |
| CosPlaceBackbone=ResNet-502024.05 | 99.6 | — | — | |
| MixVPRBackbone=ResNet-502024.05 | 99.6 | — | — | |
| EigenPlacesBackbone=ResNet-502024.05 | 99.6 | — | — | |
| EigenPlaces2025.04 | 99.6 | 99.9 | 100 | |
| CosPlaceLearning Protocol=Supervised Learning2026.01 | 99.6 | 99.9 | — | |
| EigenPlacesBackbone=ResNet-50, Desc. Dim.=5122023.08 | 99.5 | — | — | |
| ProGeo (Transformer)Descriptor dimension=512, Train set=SF-XL2024.06 | 99.5 | 99.9 | — | |
| SelaVPR(global)reranking=false2024.02 | 99.4 | 99.9 | 100 | |
| Conv-APBackbone=ResNet-50, Desc. Dim.=20482023.08 | 99.3 | — | — | |
| Conv-APBackbone=ResNet-50, Desc. Dim.=5122023.08 | 99.2 | — | — | |
| MixVPRBackbone=ResNet-50, Desc. Dim.=5122023.08 | 99.2 | — | — | |
| Conv-APDescriptor dimension=512, Train set=GSV-Cities2024.06 | 99.2 | — | — | |
| MixVPRDescriptor dimension=512, Train set=GSV-Cities2024.06 | 99.2 | — | — | |
| CosPlaceBackbone=ResNet-50, Desc. Dim.=5122023.08 | 99.1 | — | — | |
| MixVPRBackbone=ResNet-50, Desc. Dim.=1282023.08 | 99 | — | — | |
| EigenPlacesBackbone=ResNet-50, Desc. Dim.=1282023.08 | 98.8 | — | — | |
| CosPlaceBackbone=ResNet-50, Desc. Dim.=1282023.08 | 98.7 | — | — | |
| TransVPR2024.02 | 98.7 | 99 | 99.2 | |
| DINOv2-RIALearning Protocol=training-free (with pre-trained backbone)2026.01 | 97.2 | 98.7 | — | |
| DINOv2-VLADLearning Protocol=Unsupervised Learning (with pre-trained backbone)2026.01 | 96.5 | 99.1 | — | |
| AnyLoc-VLAD-DINOv2-PCALearning Protocol=Unsupervised Learning (with pre-trained backbone)2026.01 | 96.4 | 99.5 | — | |
| EigenPlaces2024.05 | 95.6 | 99.9 | 100 | |
| EigenPlacesBackbone=VGG-16, Desc. Dim.=5122023.08 | 95.4 | — | — | |
| EigenPlacesBackbone=VGG-162024.05 | 95.4 | — | — | |
| CosPlaceBackbone=VGG-16, Desc. Dim.=5122023.08 | 95.3 | — | — | |
| CosPlaceBackbone=VGG-162024.05 | 95.3 | — | — | |
| CosPlaceDesc. dim.=512, Train set=SF-XL, Backbone=VGG-162022.04 | 94.3 | 97.4 | — | |
| CosPlaceDescriptor dimension=512, Train set=SF-XL2024.06 | 94.3 | 97.4 | — | |
| Patch-NetVLAD-pvariant=performance2024.02 | 93.9 | 95.5 | 96.2 | |
| NetVLADDesc. dim.=32768, Train set=MSLS, Backbone=VGG-162022.04 | 92.8 | 97.6 | — | |
| NetVLADDescriptor dimension=32768, Train set=MSLS2024.06 | 92.8 | 97.6 | — | |
| SelaVPR2024.05 | 92.5 | 100 | 100 | |
| Patch-NetVLAD-svariant=speed2024.02 | 90.2 | 93.6 | 95 | |
| CricaVPR2024.05 | 90 | 99.9 | 99.9 | |
| DINOv2-GeMLearning Protocol=training-free (with pre-trained backbone)2026.01 | 89 | 95.6 | — | |
| AnyLoc-VLAD-DINO-PCALearning Protocol=Unsupervised Learning (with pre-trained backbone)2026.01 | 88.5 | 94.9 | — | |
| SP-SuperGlue2024.02 | 86.5 | 92.1 | 93.4 | |
| GeMDesc. dim.=512, Train set=SF-XL*, Backbone=VGG-162022.04 | 84.6 | 93.3 | — | |
| GeMDescriptor dimension=512, Train set=MSLS2024.06 | 84.6 | 93.3 | — | |
| NetVLADDesc. dim.=32768, Train set=SF-XL*, Backbone=VGG-162022.04 | 80.6 | 90.7 | — | |
| NetVLADDesc. dim.=32768, Train set=SF-XL**, Backbone=VGG-162022.04 | 78.5 | 88.3 | — | |
| SFRSBackbone=VGG-16, Desc. Dim.=40962023.08 | 75.9 | — | — | |
| SFRSBackbone=VGG-162024.05 | 75.9 | — | — | |
| SAREDesc. dim.=4096, Train set=Pitts30k, Backbone=VGG-162022.04 | 72.7 | 86 | — | |
| SAREDescriptor dimension=4096, Train set=Pitts30k2024.06 | 72.7 | 86 | — | |
| SFRSDesc. dim.=4096, Train set=Pitts30k, Backbone=VGG-162022.04 | 72.5 | 85.4 | — | |
| SFRSDescriptor dimension=4096, Train set=Pitts30k2024.06 | 72.5 | 85.4 | — | |
| CRNDesc. dim.=32768, Train set=Pitts30k, Backbone=VGG-162022.04 | 70.9 | 82.8 | — | |
| CRNDescriptor dimension=32768, Train set=Pitts30k2024.06 | 70.9 | 82.8 | — | |
| NetVLADDesc. dim.=32768, Train set=Pitts30k, Backbone=VGG-162022.04 | 70.8 | 81.8 | — | |
| NetVLADDescriptor dimension=32768, Train set=Pitts30k2024.06 | 70.8 | 81.8 | — | |
| GeMDesc. dim.=512, Train set=MSLS, Backbone=VGG-162022.04 | 68.5 | 82.7 | — | |
| GeMDesc. dim.=512, Train set=Pitts30k, Backbone=VGG-162022.04 | 66.7 | 78.9 | — | |
| NetVLADBackbone=VGG-16, Desc. Dim.=40962023.08 | 64.6 | — | — | |
| NetVLADBackbone=VGG-162024.05 | 64.6 | — | — | |
| GeMDescriptor dimension=512, Train set=Pitts30k2024.06 | 59.9 | 76.3 | — | |
| NetVLADLearning Protocol=Supervised Learning2026.01 | 57.9 | 73 | — |