3D Object Detection on nuScenes frontal (val)
0.71MAE (0-20m)MonoMAE
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| MonoMAEZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.71 | 1.57 | 4.95 | 1.4 | |
| MonoUNIZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.72 | 1.79 | 4.98 | 1.43 | |
| DEVIANTcross-dataset=trained on KITTI2022.07 | 0.76 | 1.6 | 4.5 | 1.26 | |
| DEVIANTZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.76 | 1.6 | 4.5 | 1.26 | |
| GUP Netcross-dataset=trained on KITTI2022.07 | 0.82 | 1.7 | 6.2 | 1.45 | |
| GUPNetZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.82 | 1.7 | 6.2 | 1.45 | |
| M3D-RPNcross-dataset=trained on KITTI2022.07 | 0.94 | 3.06 | 10.36 | 2.67 | |
| MonoRCNNcross-dataset=trained on KITTI2022.07 | 0.94 | 2.84 | 8.65 | 2.39 | |
| M3D-RPNZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.94 | 3.06 | 10.36 | 2.67 | |
| MonoRCNNZero-shot generalization=true, Pre-trained on KITTI=true2024.05 | 0.94 | 2.84 | 8.65 | 2.39 |