Monocular Depth Estimation on Waymo Open dataset
0.095AbsRelViT-22B
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ViT-22BBackbone=ViT-22B, Decoder=DPT2023.02 | 0.095 | 0.021 | 70.2 | 90.9 | 97.9 | — | — | — | — | |
| ViT-eBackbone=ViT-e, Decoder=DPT2023.02 | 0.112 | 0.024 | 63.1 | 88.8 | 97.5 | — | — | — | — | |
| Dynamo-DepthYear=2023, Resolution (pixels)=320 x 4802024.05 | 0.116 | — | — | 87.8 | 96.9 | 1.156 | 6 | 0.166 | 98.9 | |
| DCPI-DepthYear=-, Resolution (pixels)=320 x 4802024.05 | 0.116 | — | — | 87.2 | 97.2 | 0.963 | 5.642 | 0.162 | 99.1 | |
| ViT-LBackbone=ViT-L, Decoder=DPT2023.02 | 0.121 | 0.027 | 59.4 | 87.1 | 97.2 | — | — | — | — | |
| DIFFNetYear=2021, Resolution (pixels)=320 x 4802024.05 | 0.149 | — | — | 83.8 | 95.6 | 2.082 | 7.474 | 0.2 | 98.1 | |
| Lite-Mono-8MYear=2023, Resolution (pixels)=320 x 4802024.05 | 0.154 | — | — | 82.5 | 94.7 | 2.297 | 7.495 | 0.209 | 97.5 | |
| Li et al.Year=2021, Resolution (pixels)=320 x 4802024.05 | 0.157 | — | — | — | — | 1.531 | 7.09 | 0.205 | — | |
| Lite-MonoYear=2023, Resolution (pixels)=320 x 4802024.05 | 0.158 | — | — | 81.6 | 94.4 | 2.305 | 7.394 | 0.21 | 97.6 | |
| ViT-22BBackbone=ViT-22B, Decoder=Linear2023.02 | 0.166 | 0.039 | 41.2 | 77.9 | 96 | — | — | — | — | |
| Monodepth2Year=2019, Resolution (pixels)=320 x 4802024.05 | 0.173 | — | — | 79.7 | 93 | 2.731 | 7.708 | 0.227 | 96.8 | |
| π³input_frames=62026.02 | 0.19 | — | — | — | — | — | 6.68 | — | — | |
| ViT-eBackbone=ViT-e, Decoder=Linear2023.02 | 0.204 | 0.053 | 33.2 | 68.7 | 93.8 | — | — | — | — | |
| LFGinput_frames=32026.02 | 0.21 | — | — | — | — | — | 6.87 | — | — | |
| ViT-LBackbone=ViT-L, Decoder=Linear2023.02 | 0.222 | 0.06 | 30.4 | 65.2 | 92.6 | — | — | — | — |