3D Semantic Segmentation on vKITTI (6-fold val)
85.1OA (Overall Accuracy)MuGNet
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MuGNetInput Features=XYZ coordinates without RGB values2020.09 | 85.1 | 50 | 70 | 88.6 | 35.2 | 63 | 80.2 | 40.8 | 32 | 56.3 | 23.4 | 3.92 | 7.1 | 84.3 | 65.4 | |
| Engelmann et al.Input Features=XYZ coordinates without RGB values2020.09 | 73.2 | 26.4 | 38.9 | 87.1 | 14.6 | 44 | 58.4 | 12.4 | 9.4 | 10.6 | 5.3 | 2.2 | 3.6 | 43 | 13.3 | |
| PointNetInput Features=XYZ coordinates without RGB values2020.09 | 63.3 | 17.9 | 32.9 | 76.4 | 11.9 | 11.7 | 49.9 | 3.6 | 2.8 | 3.7 | 3.5 | 0.7 | 1.5 | 25.1 | 3.4 |