Camera Pose Regression on Cambridge Landmarks (test)
0.31Translation Error (Kings College, Median, m)PMNet
Evaluation results
| Method | Links | ||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PMNetRegression type=APR, Dataset-specific training time=Days / scene2025.12 | 0.31 | — | — | — | — | — | — | — | — | — | — | — | — | 0.55 | 0.44 | 0.79 | 0.17 | 0.86 | 0.31 | 0.96 | 0.31 | 0.79 | — | — | |
| LENSRegression type=APR, Dataset-specific training time=Days / scene2025.12 | 0.33 | — | — | — | — | — | — | — | — | — | — | — | — | 0.5 | 0.44 | 0.9 | 0.27 | 1.6 | 0.53 | 1.6 | 0.39 | 1.2 | — | — | |
| Reloc-VGGTRegression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 0.36 | — | — | — | — | — | — | — | — | — | — | 0.59 | 0.43 | 0.36 | 0.57 | 0.44 | 0.1 | 0.33 | 0.24 | 0.43 | 0.32 | 0.37 | 0.37 | 0.38 | |
| DFNet+NeFeSRegression type=APR, Dataset-specific training time=Days / scene2025.12 | 0.37 | — | — | — | — | — | — | — | — | — | — | — | — | 0.54 | 0.52 | 0.88 | 0.15 | 0.53 | 0.37 | 1.14 | 0.35 | 0.77 | — | — | |
| Reloc-VGGT(mask)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 0.4 | — | — | — | — | — | — | — | — | — | — | 0.98 | 0.63 | 0.48 | 0.59 | 0.52 | 0.14 | 0.56 | 0.31 | 0.55 | 0.36 | 0.55 | 0.48 | 0.55 | |
| Reloc3r-512Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 0.42 | — | — | — | — | — | — | — | — | — | — | 1.22 | 0.73 | 0.36 | 0.62 | 0.55 | 0.13 | 0.58 | 0.34 | 0.58 | 0.38 | 0.52 | 0.55 | 0.56 | |
| EssNet (CL)Regression type=RPR, Training protocol=Seen, Dataset-specific training time=Hours2025.12 | 0.48 | — | — | — | — | — | — | — | — | — | — | 3.2 | 2.2 | 1 | 1.14 | 2.5 | 0.48 | 2.5 | 1.52 | 3.2 | 1.08 | 3.41 | 1.37 | 2.3 | |
| AnchorNetRegression type=RPR, Training protocol=Seen, Dataset-specific training time=Hours2025.12 | 0.57 | — | — | — | — | — | — | — | — | — | — | — | — | 0.88 | 1.21 | 2.55 | 0.52 | 2.27 | 1.04 | 2.69 | 0.84 | 2.1 | — | — | |
| Cov2Pose2026.03 | 1.57 | — | — | — | — | — | — | — | — | — | — | — | — | 1.81 | 2.08 | 2.35 | 1.11 | 5.33 | 2.06 | 5 | — | — | 1.7 | 3.56 | |
| GPoseNet2026.03 | 1.61 | — | — | — | — | — | — | — | — | — | — | — | — | 2.29 | 2.62 | 3.89 | 1.14 | 5.73 | 2.93 | 6.46 | — | — | 2.07 | 4.59 | |
| BayesianPN2026.03 | 1.74 | — | — | — | — | — | — | — | — | — | — | — | — | 4.06 | 2.57 | 5.14 | 1.25 | 7.54 | 2.11 | 8.38 | — | — | 1.91 | 6.28 | |
| PoseNet2026.03 | 1.92 | — | — | — | — | — | — | — | — | — | — | — | — | 5.4 | 2.31 | 5.38 | 1.46 | 8.08 | 2.65 | 8.48 | — | — | 2.08 | 6.83 | |
| ExReNet (SUNCG)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 2.33 | — | — | — | — | — | — | — | — | — | — | 9.79 | 4.46 | 2.48 | 3.54 | 3.49 | 0.72 | 2.41 | 2.3 | 3.74 | 2.22 | 3.03 | 3.74 | 3.31 | |
| ExReNet (SN)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 2.34 | — | — | — | — | — | — | — | — | — | — | 10.97 | 6.52 | 2.99 | 3.4 | 4.92 | 0.9 | 3.23 | 2.6 | 4.69 | 2.36 | 4.23 | 4.08 | 4.32 | |
| Map-free (Regress)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 2.44 | — | — | — | — | — | — | — | — | — | — | 8.4 | 4.56 | 2.54 | 3.73 | 5.23 | 0.97 | 3.17 | 2.91 | 5.1 | 2.51 | 4.01 | 3.69 | 4.12 | |
| Map-free (Match)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | 2.51 | — | — | — | — | — | — | — | — | — | — | 9.09 | 5.33 | 3.11 | 3.89 | 6.44 | 1.04 | 3.61 | 3 | 6.14 | 2.61 | 4.83 | 3.9 | 4.93 | |
| AD-PoseNetBackbone=ResNet34, input=12021.03 | — | — | — | 1.3 | 1.67 | 2.28 | 4.8 | 1.22 | 4.64 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Bay. PoseNetinput=12021.03 | — | 2.57 | 5.14 | 1.74 | 4.06 | 2.11 | 8.38 | 1.25 | 7.54 | 1.92 | 6.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoordiNetBackbone=EffNet b3, input=12021.03 | — | 0.97 | 2.08 | 0.7 | 0.92 | 1.32 | 3.56 | 0.69 | 3.74 | 0.92 | 2.58 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoordiNetBackbone=ResNet34, input=12021.03 | — | 1.43 | 2.86 | 0.8 | 1.22 | 1.32 | 4.1 | 0.73 | 4.69 | 1.07 | 3.22 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EssNet (7S)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 10.36 | 85.75 | — | — | |
| ImageNet+NCMRegression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 1.83 | 0.56 | — | — | |
| LSTM-Poseinput=22021.03 | — | 1.51 | 4.29 | 0.99 | 3.65 | 1.52 | 6.68 | 1.18 | 7.44 | 1.3 | 5.51 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MapNetBackbone=ResNet34, input=22021.03 | — | 1.94 | 3.91 | 1.07 | 1.89 | 2 | 4.53 | 1.63 | 4.22 | 1.66 | 3.64 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NC-EssNet (7S)Regression type=RPR, Training protocol=Unseen, Dataset-specific training time=None2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 9.78 | 24.35 | — | — | |
| PoseNetBackbone=ResNet34, input=12021.03 | — | 2.17 | 2.94 | 0.99 | 1.06 | 1.49 | 3.43 | 1.05 | 3.97 | 1.43 | 2.85 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Relpose-GNNRegression type=RPR, Training protocol=Seen, Dataset-specific training time=Hours2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 0.91 | 2.3 | 0.85 | 2.82 | |
| VLocNetBackbone=ResNet50, input=22021.03 | — | 1.07 | 2.41 | 0.84 | 1.42 | 0.63 | 3.91 | 0.59 | 3.53 | 0.78 | 2.82 | — | — | — | — | — | — | — | — | — | — | — | — | — |