Monocular Depth Estimation on KITTI official (test)
97.7Delta 1 AccMIM
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| MIMBackbone=SwinV2-L, Pre-training type=Supervised2023.06 | 97.7 | 99.8 | 100 | 0.05 | 0.139 | 1.966 | 0.075 | |
| PixelFormerBackbone=Swin-Large, Pre-training type=Supervised2023.06 | 97.6 | 99.7 | 99.9 | 0.051 | 0.149 | 2.081 | 0.077 | |
| BinsFormerBackbone=Swin-Large, Pre-training type=Supervised2023.06 | 97.4 | 99.7 | 99.9 | 0.052 | 0.151 | 2.098 | 0.079 | |
| DDVMBackbone=Efficient U-Net, Number of samples=2, Pre-training type=Unsupervised, Use of auxiliary supervised depth data=true2023.06 | 96.5 | 99.4 | 99.8 | 0.055 | 0.325 | 2.66 | 0.09 | |
| DDVMBackbone=Efficient U-Net, Number of samples=4, Pre-training type=Unsupervised, Use of auxiliary supervised depth data=true2023.06 | 96.5 | 99.4 | 99.8 | 0.055 | 0.292 | 2.613 | 0.089 | |
| AdaBinsBackbone=EfficientNet-B5+mini-ViT, Pre-training type=Supervised2023.06 | 96.4 | 99.5 | 99.9 | 0.058 | 0.19 | 2.36 | 0.088 | |
| DDVMBackbone=Efficient U-Net, Number of samples=1, Pre-training type=Unsupervised, Use of auxiliary supervised depth data=true2023.06 | 96.4 | 99.4 | 99.8 | 0.056 | 0.339 | 2.7 | 0.091 | |
| DPTBackbone=ResNet-50+ViT-B, Pre-training type=Supervised, Use of auxiliary supervised depth data=true2023.06 | 95.9 | 99.5 | 99.9 | 0.062 | — | 2.573 | 0.092 | |
| BTSBackbone=DenseNet-161+2023.06 | 95.6 | 99.3 | 99.8 | 0.059 | 0.245 | 2.756 | 0.096 | |
| TransDepthBackbone=ResNet-50+ViT-B, Pre-training type=Unsupervised2023.06 | 95.6 | 99.4 | 99.9 | 0.064 | 0.252 | 2.755 | 0.098 |