Monocular Depth Estimation on Cityscapes dynamic object regions (test)
0.091Abs RelFlex-X-Large
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Flex-X-LargePix.Mot.=-, Pseu.Dep.=-, Frames=1, FLOPs(G)=19.62026.07 | 0.091 | — | — | — | — | 93.5 | |
| DSI-MonoVitPix.Mot.=-, Pseu.Dep.=-, Frames=1, FLOPs(G)=47.72026.07 | 0.097 | — | — | — | — | 91.8 | |
| Ours-MonoViTResolutions=192 x 6402023.12 | 0.098 | 0.674 | 3.688 | 0.135 | 91.4 | — | |
| Ours-MonoViTResolutions=128 x 4162023.12 | 0.109 | 0.888 | 4.243 | 0.151 | 89.8 | — | |
| ProDepthPix.Mot.=-, Pseu.Dep.=-, Frames=2, FLOPs(G)=35.52026.07 | 0.115 | — | — | — | — | 88.4 | |
| Ours-CADepthResolutions=192 x 6402023.12 | 0.116 | 1.161 | 4.37 | 0.154 | 89.2 | — | |
| Ours-HR-DepthResolutions=192 x 6402023.12 | 0.117 | 1.015 | 4.297 | 0.154 | 88.3 | — | |
| Ours-Monodepth2Resolutions=192 x 6402023.12 | 0.119 | 1.044 | 4.27 | 0.157 | 88.1 | — | |
| Mining-BrNetPix.Mot.=✓, Pseu.Dep.=✓, Frames=1, FLOPs(G)=27.02026.07 | 0.119 | — | — | — | — | 87.2 | |
| DynamicDepthResolutions=128 x 4162023.12 | 0.129 | 1.273 | 4.626 | 0.168 | 86.2 | — | |
| Ours-Monodepth2Resolutions=128 x 4162023.12 | 0.136 | 1.238 | 4.791 | 0.176 | 86.4 | — | |
| InstaDMResolutions=256 x 8322023.12 | 0.139 | 1.698 | 5.76 | 0.181 | 85.9 | — | |
| MonoViTResolutions=128 x 4162023.12 | 0.142 | 1.536 | 5.13 | 0.185 | 82 | — | |
| CADepthResolutions=192 x 6402023.12 | 0.143 | 1.278 | 4.997 | 0.184 | 82.5 | — | |
| MonoViTResolutions=192 x 6402023.12 | 0.149 | 1.508 | 5.34 | 0.19 | 81.7 | — | |
| Monodepth2Resolutions=128 x 4162023.12 | 0.159 | 1.937 | 6.363 | 0.201 | 81.6 | — | |
| HR-DepthResolutions=192 x 6402023.12 | 0.165 | 2.144 | 5.72 | 0.198 | 81.1 | — | |
| Monodepth2Resolutions=192 x 6402023.12 | 0.185 | 2.432 | 5.919 | 0.218 | 79.4 | — |