Monocular Depth Estimation On Nyu Depth V2 4
Métriques
Absolute relative error (AbsRel)
Root mean square error (RMSE)
delta_1
delta_2
delta_3
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Absolute relative error (AbsRel) | Root mean square error (RMSE) | delta_1 | delta_2 | delta_3 |
|---|---|---|---|---|---|
| moving-indoor-unsupervised-video-depth | 0.208 | 0.712 | 67.4 | 90.0 | 96.8 |
| structdepth-leveraging-the-structural | 0.142 | 0.540 | 81.3 | 95.4 | 98.8 |
| deeper-into-self-supervised-monocular-indoor | 0.126 | 0.494 | 84.5 | 96.5 | 99.1 |
| toward-practical-self-supervised-monocular | 0.130 | 0.517 | 83.2 | 96.3 | 99.0 |
| monoindoor-towards-good-practice-of-self | 0.134 | 0.526 | 82.3 | 95.8 | 98.9 |
| towards-better-generalization-joint-depth | 0.189 | 0.686 | 70.1 | 91.2 | 97.8 |
| unsupervised-scale-consistent-depth-learning | 0.157 | 0.593 | 78.0 | 94.0 | 98.4 |
| p-2-net-patch-match-and-plane-regularization | 0.147 | 0.553 | 80.4 | 95.2 | 98.7 |