Monocular Depth Estimation on KITTI 14 (Eigen split)
0.071Abs RelSPIDepth
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SPIDepthTraining Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.071 | 0.531 | 3.662 | 0.153 | 94 | 97.3 | 98.5 | |
| SQLdepthTraining Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.075 | 0.539 | 3.722 | 0.156 | 93.7 | 97.3 | 98.5 | |
| ManyDepthTraining Mode=M (Monocular videos), Test Setting=2(-1, 0)+TTR (Two frames + Test-Time Refinement), Input Resolution=1024 x 3202024.04 | 0.087 | 0.685 | 4.142 | 0.167 | 92 | 96.8 | 98.3 | |
| EPCDepthTraining Mode=S+Distill (Stereo + self-distillation), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.091 | 0.646 | 4.207 | 0.176 | 90.1 | 96.6 | 98.3 | |
| Depth HintsTraining Mode=S+Aux (Stereo + auxiliary information), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.096 | 0.71 | 4.393 | 0.185 | 89 | 96.2 | 98.1 | |
| CADepth-NetTraining Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.096 | 0.694 | 4.264 | 0.173 | 90.8 | 96.8 | 98.4 | |
| DIFFNetTraining Mode=M (Monocular videos), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.097 | 0.722 | 4.345 | 0.174 | 90.7 | 96.7 | 98.4 | |
| FeatDepth-MSTraining Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.099 | 0.697 | 4.427 | 0.184 | 88.9 | 96.3 | 98.2 | |
| HR-DepthTraining Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.101 | 0.716 | 4.395 | 0.179 | 89.9 | 96.6 | 98.3 | |
| XDistillTraining Mode=S+Distill (Stereo + self-distillation), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.102 | 0.698 | 4.439 | 0.18 | 89.5 | 96.5 | 98.3 | |
| Monodepth2Training Mode=MS (Monocular and Stereo pairs), Test Setting=1 (One single frame input), Input Resolution=1024 x 3202024.04 | 0.106 | 0.806 | 4.63 | 0.193 | 87.6 | 95.8 | 98 | |
| Wang et al.Training Mode=M (Monocular videos), Test Setting=2(-1, 0) (Two frames: previous and current), Input Resolution=1024 x 3202024.04 | 0.106 | 0.773 | 4.491 | 0.185 | 89 | 96.2 | 98.2 |