Visual Place Recognition on YQ360
0.7Query Time (ms)SwinT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SwinTBackbone=Swin-T, #Params. (M)=28.292025.06 | 0.7 | 27.4 | 63.5 | 72.3 | |
| ConvNeXtSBackbone=ConvNeXt-S, #Params. (M)=50.222025.06 | 0.7 | 37 | 72.5 | 83.9 | |
| Berton et al. (ViT-B)Backbone=ViT-B, #Params. (M)=86.862025.06 | 0.7 | 40.4 | 74.8 | 88.4 | |
| HypeVPR-OBackbone=ResNet-50, #Params. (M)=27.70, trained_on_extra_large_dataset=true2025.06 | 1.9 | 53.6 | 81.2 | 88.8 | |
| HypeVPR-BBackbone=Swin-T, #Params. (M)=28.292025.06 | 4 | 38 | 74.4 | 88 | |
| HypeVPR-BBackbone=ConvNeXt-S, #Params. (M)=50.222025.06 | 4 | 43.8 | 78.8 | 89.2 | |
| PanoVPRBackbone=Swin-T, #Params. (M)=28.29, sliding_window_crops=82025.06 | 4.9 | 30.8 | 69.6 | 81.6 | |
| PanoVPRBackbone=ConvNeXt-S, #Params. (M)=50.22, sliding_window_crops=82025.06 | 4.9 | 39.6 | 76.8 | 87.6 | |
| HypeVPR-LBackbone=Swin-T, #Params. (M)=28.292025.06 | 6.7 | 45.6 | 84 | 91.2 | |
| HypeVPR-LBackbone=ConvNeXt-S, #Params. (M)=50.222025.06 | 6.7 | 52.4 | 85.2 | 94.8 | |
| PanoVPRBackbone=Swin-T, #Params. (M)=28.29, sliding_window_crops=162025.06 | 11 | 43.2 | 82.4 | 90.8 | |
| PanoVPRBackbone=ConvNeXt-S, #Params. (M)=50.22, sliding_window_crops=162025.06 | 11 | 46 | 83.2 | 92.4 | |
| HypeVPR-BBackbone=ResNet-50, #Params. (M)=27.70, trained_on_extra_large_dataset=true2025.06 | 14.2 | 63.6 | 88.8 | 96.4 | |
| NetVLADBackbone=ResNet-50, #Params. (M)=25.302025.06 | 19.1 | 35.2 | 66.8 | 80 | |
| Orhan et al.Backbone=ResNet-101, #Params. (M)=136.62, trained_on_extra_large_dataset=true2025.06 | 981.4 | 47.6 | 79.2 | 88.4 |