Monocular Depth Estimation on KITTI capped 50m 15 (Eigen)
0.1069Abs RelLearn Stereo, Infer Mono (ResNet)
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Learn Stereo, Infer Mono (ResNet)Training Dataset=CS + K, Backbone=ResNet, Post-processing=true2019.05 | 0.1069 | 0.6531 | 3.79 | 0.195 | 86.7 | 95.4 | 97.9 | |
| Godard et al.Training Dataset=CS + K2019.05 | 0.108 | 0.657 | 3.729 | 0.194 | 87.3 | 95.4 | 97.9 | |
| Learn Stereo, Infer Mono (VGG)Training Dataset=CS + K, Backbone=VGG, Post-processing=true2019.05 | 0.1155 | 0.7152 | 3.922 | 0.201 | 85.9 | 95.1 | 97.9 | |
| Kuznietsov et al.Supervised=Yes, Training Dataset=K, Cap=50m2019.04 | 0.117 | 0.597 | 3.531 | 0.183 | 86.1 | 96.4 | 98.9 | |
| Godard et al.Supervised=No, Training Dataset=K, Cap=50m2019.04 | 0.14 | 0.976 | 4.471 | 0.232 | 81.8 | 93.1 | 96.9 | |
| GASDASupervised=No, Training Dataset=K+S(DA), Cap=50m2019.04 | 0.143 | 0.756 | 3.846 | 0.217 | 83.6 | 94.6 | 97.6 | |
| Yin et al.Training Dataset=K2019.05 | 0.147 | 0.936 | 4.348 | 0.218 | 81 | 94.1 | 97.7 | |
| All real (baseline2)Supervised=No, Training Dataset=K, Cap=50m2019.04 | 0.151 | 0.856 | 4.043 | 0.227 | 82.4 | 94 | 97.3 | |
| Kundu et al.Supervised=Semi, Training Dataset=K+S(DA), Cap=50m2019.04 | 0.162 | 1.041 | 4.344 | 0.225 | 78.4 | 93 | 97.4 | |
| Zheng et al.Supervised=No, Training Dataset=K+S(DA), Cap=50m2019.04 | 0.168 | 1.199 | 4.674 | 0.243 | 77.2 | 91.2 | 96.6 | |
| Garg et al.Training Dataset=KITTI, Pose Supervision=true, Evaluation Cap=50m2017.04 | 0.169 | 1.08 | 5.104 | 0.273 | 74 | 90.4 | 96.2 | |
| Garg et al.Supervised=No, Training Dataset=K, Cap=50m2019.04 | 0.169 | 1.08 | 5.104 | 0.273 | 74 | 90.4 | 96.2 | |
| Garg et al.Training Dataset=K2019.05 | 0.169 | 1.08 | 5.104 | 0.273 | 74 | 90.4 | 96.2 | |
| SfMLearnerTraining Dataset=Cityscapes + KITTI, Explainability=true, Evaluation Cap=50m2017.04 | 0.19 | 1.436 | 4.975 | 0.258 | 73.5 | 91.5 | 96.8 | |
| SfMLearnerTraining Dataset=KITTI, Explainability=true, Evaluation Cap=50m2017.04 | 0.201 | 1.391 | 5.181 | 0.264 | 69.6 | 90 | 96.6 | |
| Kundu et al.Supervised=No, Training Dataset=K+S(DA), Cap=50m2019.04 | 0.203 | 1.734 | 6.251 | 0.284 | 68.7 | 89.9 | 95.8 | |
| SfMLearner (w/o explainability)Training Dataset=KITTI, Explainability=false, Evaluation Cap=50m2017.04 | 0.208 | 1.551 | 5.452 | 0.273 | 69.5 | 90 | 96.4 | |
| All synthetic (baseline1)Supervised=No, Training Dataset=S, Cap=50m2019.04 | 0.244 | 1.771 | 5.354 | 0.313 | 64.7 | 86.6 | 94.3 | |
| SfMLearnerTraining Dataset=Cityscapes, Explainability=true, Evaluation Cap=50m2017.04 | 0.26 | 2.232 | 6.148 | 0.321 | 59 | 85.2 | 94.5 |