Visual Localization on Extended CMU Seasons Urban
96.9Recall @ (0.25m, 2°)HLoc+PixLoc
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| HLoc+PixLocMethod Group=S2024.02 | 96.9 | 98.9 | 99.3 | |
| DeViLocMethod Group=SD2024.02 | 95.7 | 98.4 | 99.2 | |
| HLoc[SP+SG]Method Group=S2024.02 | 95.5 | 98.6 | 99.3 | |
| PFSLBackbone=FGSN, Clusters=200, Training dataset=CMU2019.08 | 95.3 | 99.5 | 100 | |
| D2NetMethod Group=S2024.02 | 94 | 97.7 | 99.1 | |
| HF-Net2019.08 | 89.5 | 94.2 | 97.9 | |
| HLoc[SP]Method Group=S2024.02 | 89.5 | 94.2 | 97.9 | |
| PixLocMethod Group=S2024.02 | 88.3 | 90.4 | 93.7 | |
| GSMCBackbone=FGSN, Clusters=200, Training dataset=CMU2019.08 | 86.4 | 91.2 | 93.8 | |
| SSMCBackbone=FGSN, Clusters=100, Training dataset=CMU2019.08 | 85.3 | 91 | 94.6 | |
| PFSL2019.08 | 84.7 | 96.8 | 100 | |
| GSMC2019.08 | 84.3 | 89.4 | 93.2 | |
| ASMethod Group=S2024.02 | 81 | 87.3 | 92.4 | |
| City Scale Localization2019.08 | 71.2 | 74.6 | 78.7 | |
| Asymmetric Hypercolumn Matching2019.08 | 65.7 | 82.7 | 91 | |
| DenseVLAD2019.08 | 14.7 | 36.3 | 83.9 | |
| NetVLAD2019.08 | 12.2 | 31.5 | 89.8 | |
| Patch NetVLADPCA=Yes, Dimension=4096, Re-ranking=true, Training Dataset=MSLS2023.03 | 11.8 | 36.2 | 96.2 | |
| SP-SuperGlueRe-ranking=true, Training Dataset=MSLS2023.03 | 9.5 | 30.7 | 96.7 | |
| ResNeXt-GeM-GCLPCA=Yes, Dimension=1024, Re-ranking=false, Training Dataset=MSLS2023.03 | 6.1 | 18.2 | 74.9 | |
| DELGRe-ranking=true, Training Dataset=MSLS2023.03 | 5.7 | 21.1 | 93.6 | |
| VGG-GeM-GCLPCA=Yes, Dimension=512, Re-ranking=false, Training Dataset=MSLS2023.03 | 5.7 | 17.1 | 66.3 | |
| ResNet50-GeM-GCLPCA=Yes, Dimension=1024, Re-ranking=false, Training Dataset=MSLS2023.03 | 5.4 | 16.5 | 69.9 | |
| ResNet152-GeM-GCLPCA=Yes, Dimension=2048, Re-ranking=false, Training Dataset=MSLS2023.03 | 5.3 | 16.1 | 66.4 | |
| NetVLAD 16PCA=Yes, Dimension=4096, Re-ranking=false, Training Dataset=MSLS2023.03 | 4.4 | 13.7 | 61.4 | |
| NetVLAD 64PCA=Yes, Dimension=4096, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.9 | 12.1 | 58.4 | |
| ResNet50-GeM-GCLPCA=No, Dimension=2048, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.8 | 11.8 | 61.6 | |
| VGG-GeM-GCLPCA=No, Dimension=512, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.6 | 11.2 | 55.8 | |
| ResNet152-GeM-GCLPCA=No, Dimension=2048, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.6 | 11.3 | 63.1 | |
| ResNeXt-GeM-GCLPCA=No, Dimension=2048, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.5 | 10.5 | 58.8 | |
| NetVLAD-GCLPCA=Yes, Dimension=4096, Re-ranking=false, Training Dataset=MSLS2023.03 | 3.1 | 9.7 | 52.4 | |
| NetVLAD-GCLPCA=No, Dimension=32768, Re-ranking=false, Training Dataset=MSLS2023.03 | 3 | 9.7 | 52.3 | |
| NetVLAD 16PCA=No, Dimension=8192, Re-ranking=false, Training Dataset=MSLS2023.03 | 1.7 | 5.5 | 39.1 | |
| NetVLAD 64PCA=No, Dimension=32768, Re-ranking=false, Training Dataset=MSLS2023.03 | 1.3 | 4.5 | 31.9 |