Monocular Depth Estimation on DCM 300 image-depth pairs (val)
0.251AbsRelComicsDepth
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ComicsDepthTranslation Model=DUNIT, Architecture=contextual depth estimator with feature-based GAN, Laplacian and text detection module2021.10 | 0.251 | 0.318 | 0.971 | 0.305 | |
| CDETranslation Model=DUNIT2021.10 | 0.304 | 0.374 | 1.024 | 0.367 | |
| MIDASTranslation Model=DUNIT2021.10 | 0.309 | 0.381 | 1.033 | 0.375 | |
| Song et al.Translation Model=DUNIT2021.10 | 0.339 | 0.401 | 1.098 | 0.402 | |
| T2NetTranslation Model=DUNIT2021.10 | 0.351 | 0.416 | 1.117 | 0.415 |