Robust 3D Semantic Segmentation On Nuscenes C
Métriques
mean Corruption Error (mCE)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | mean Corruption Error (mCE) |
|---|---|
| searching-efficient-3d-architectures-with | 106.65% |
| cylindrical-and-asymmetrical-3d-convolution | 105.56% |
| searching-efficient-3d-architectures-with | 97.45% |
| polarnet-an-improved-grid-representation-for | 115.09% |
| cylindrical-and-asymmetrical-3d-convolution | 111.84% |
| 4d-spatio-temporal-convnets-minkowski | 96.37% |
| gfnet-geometric-flow-network-for-3d-point | 92.55% |
| using-a-waffle-iron-for-automotive-point | 106.73% |
| cenet-toward-concise-and-efficient-lidar | 112.79% |
| 4d-spatio-temporal-convnets-minkowski | 100.00% |
| fidnet-lidar-point-cloud-semantic | 122.42% |
| 2dpass-2d-priors-assisted-semantic | 98.56% |