Monocular Depth Estimation on Cityscapes 12 (test)
0.106Abs RelSQLdepth (Fine-tuned)
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SQLdepth (Fine-tuned)Train=-, K→C, Test=1, Resolution=416 x 1282023.09 | 0.106 | 1.173 | 6.237 | 0.163 | 88.8 | 97.2 | 99 | |
| SQLdepth (From scratch)Train=-, C, Test=1, Resolution=416 x 1282023.09 | 0.11 | 1.13 | 6.264 | 0.165 | 88.1 | 97.1 | 99.1 | |
| InstaDMTrain=MMask, C, Test=1, Resolution=832 x 2562023.09 | 0.111 | 1.158 | 6.437 | 0.182 | 86.8 | 96.1 | 98.3 | |
| ManyDepthTrain=MMask, C, Test=2(-1,0), Resolution=416 x 1282023.09 | 0.114 | 1.193 | 6.223 | 0.17 | 87.5 | 96.7 | 98.9 | |
| Lee et al.Train=MMask, C, Test=1, Resolution=832 x 2562023.09 | 0.116 | 1.213 | 6.695 | 0.186 | 85.2 | 95.1 | 98.2 | |
| Li et al.Train=MMask, C, Test=1, Resolution=416 x 1282023.09 | 0.119 | 1.29 | 6.98 | 0.19 | 84.6 | 95.2 | 98.2 | |
| SQLdepth (Zero-shot)Train=-, K, Test=1, Resolution=416 x 1282023.09 | 0.125 | 1.347 | 7.398 | 0.194 | 83.4 | 95.1 | 98.5 | |
| Videos in the WildTrain=MMask, C, Test=1, Resolution=416 x 1282023.09 | 0.127 | 1.33 | 6.96 | 0.195 | 83 | 94.7 | 98.1 | |
| Monodepth2Train=-, C, Test=1, Resolution=416 x 1282023.09 | 0.129 | 1.569 | 6.876 | 0.187 | 84.9 | 95.7 | 98.3 | |
| Struct2Depth 2Train=MMask, C, Test=1, Resolution=416 x 1282023.09 | 0.145 | 1.737 | 7.28 | 0.205 | 81.3 | 94.2 | 97.6 | |
| Pilzer et al.Train=GAN, C, Test=1, Resolution=512 x 2562023.09 | 0.24 | 4.264 | 8.049 | 0.334 | 71 | 87.1 | 93.7 |