Robust 3D Semantic Segmentation On
Metrics
mean Corruption Error (mCE)
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | mean Corruption Error (mCE) |
|---|---|
| polarnet-an-improved-grid-representation-for | 118.56% |
| squeezesegv2-improved-model-structure-and | 152.45% |
| kpconv-flexible-and-deformable-convolution | 99.54% |
| 2dpass-2d-priors-assisted-semantic | 106.14% |
| 4d-spatio-temporal-convnets-minkowski | 100.00% |
| pids-joint-point-interaction-dimension-search | 104.13% |
| cylindrical-and-asymmetrical-3d-convolution | 103.13% |
| cpgnet-cascade-point-grid-fusion-network-for | 107.34% |
| pids-joint-point-interaction-dimension-search | 101.20% |
| gfnet-geometric-flow-network-for-3d-point | 108.68% |
| searching-efficient-3d-architectures-with | 99.16% |
| cenet-toward-concise-and-efficient-lidar | 103.41% |
| rpvnet-a-deep-and-efficient-range-point-voxel | 111.74% |
| using-a-waffle-iron-for-automotive-point | 109.54% |
| fidnet-lidar-point-cloud-semantic | 113.81% |
| cylindrical-and-asymmetrical-3d-convolution | 103.25% |
| 4d-spatio-temporal-convnets-minkowski | 100.61% |
| salsanext-fast-semantic-segmentation-of-lidar | 116.14% |
| squeezeseg-convolutional-neural-nets-with | 164.87% |
| rangenet-fast-and-accurate-lidar-semantic | 136.33% |
| rangenet-fast-and-accurate-lidar-semantic | 130.66% |
| searching-efficient-3d-architectures-with | 100.30% |