Monocular Depth Estimation on KITTI 2012 2015 (Eigen split)
0.091Abs Rel ErrorVOMonodepth - ResNet
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| VOMonodepth - ResNetSupervision=Stereo, Post-processing=true, Visual Odometry variant=true2019.08 | 0.091 | 0.548 | 3.69 | 0.181 | 89.2 | 95.6 | 97.9 | |
| VOPyD-NetSupervision=Stereo, Post-processing=true, Visual Odometry variant=true2019.08 | 0.102 | 0.611 | 3.81 | 0.188 | 87.6 | 95.2 | 97.9 | |
| MonoResMatchSupervision=Stereo+SGM, Post-processing=true, Visual Odometry variant=false2019.08 | 0.102 | 0.563 | 3.725 | 0.183 | 88.5 | 96.4 | 98.6 | |
| VOPyD-NetSupervision=Stereo, Post-processing=false, Visual Odometry variant=true2019.08 | 0.105 | 0.916 | 4.916 | 0.203 | 87.4 | 94.6 | 97.4 | |
| 3Net - ResNetSupervision=Stereo, Post-processing=true, Visual Odometry variant=false2019.08 | 0.106 | 0.627 | 3.982 | 0.192 | 86.9 | 95.3 | 97.9 | |
| Monodepth - ResNetSupervision=Stereo, Post-processing=true, Visual Odometry variant=false2019.08 | 0.108 | 0.679 | 4.123 | 0.194 | 86.8 | 95.2 | 97.8 | |
| PyD-NetSupervision=Stereo, Post-processing=true, Visual Odometry variant=false2019.08 | 0.123 | 0.733 | 4.333 | 0.21 | 83.4 | 94.3 | 97.6 | |
| PyD-NetSupervision=Stereo, Post-processing=false, Visual Odometry variant=false2019.08 | 0.13 | 0.833 | 4.569 | 0.219 | 82.5 | 93.8 | 97.4 | |
| DF-NetSupervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.131 | 0.706 | 4.365 | 0.196 | 83.1 | 95.2 | 98.4 | |
| LKVOSupervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.135 | 0.812 | 4.246 | 0.2 | 83.6 | 95.2 | 98.2 | |
| Struct2depth (M)Supervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.135 | 0.792 | 4.356 | 0.197 | 83.6 | 95.5 | 98.4 | |
| GeoNet - ResNetSupervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.141 | 0.842 | 4.688 | 0.209 | 81.2 | 94.5 | 98.1 | |
| Vid2depthSupervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.143 | 0.827 | 4.702 | 0.213 | 81.2 | 94.3 | 97.9 | |
| Zhan et al.Training data=KITTI [28]2019.03 | 0.144 | 1.391 | 5.869 | 0.241 | 80.3 | 92.8 | 96.9 | |
| Godard et al.Training data=KITTI [28]2019.03 | 0.148 | 1.344 | 5.927 | 0.247 | 80.3 | 92.2 | 96.4 | |
| SfmLearnerSupervision=Mono, Pre-training=CityScapes, Post-processing=false, Visual Odometry variant=false2019.08 | 0.175 | 1.309 | 5.515 | 0.247 | 74 | 91.6 | 97.1 | |
| Veritatem Dies AperitTraining data=[67]2019.03 | 0.193 | 1.438 | 5.887 | 0.234 | 83.6 | 93 | 95.8 | |
| Liu et al.Training data=KITTI [28]2019.03 | 0.202 | 1.614 | 6.523 | 0.275 | 67.8 | 89.5 | 96.5 | |
| Eigen et al.Training data=KITTI [28]2019.03 | 0.203 | 1.548 | 6.307 | 0.282 | 70.2 | 89 | 95.8 | |
| Zhou et al.Training data=KITTI [28]2019.03 | 0.208 | 1.768 | 6.856 | 0.283 | 67.8 | 88.5 | 95.7 | |
| Train Set MeanTraining data=KITTI [28]2019.03 | 0.403 | 0.53 | 8.709 | 0.403 | 59.3 | 77.6 | 87.8 |