3D Semantic Segmentation On S3Dis
Metriken
mIoU (6-Fold)
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | mIoU (6-Fold) | Paper Title | Repository |
|---|---|---|---|
| PVCNN++ | 58.98 | Point-Voxel CNN for Efficient 3D Deep Learning | |
| OneFormer3D | 75.0 | OneFormer3D: One Transformer for Unified Point Cloud Segmentation | |
| PointNext | 74.9 | PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies | |
| Superpoint Transformer | 76.0 | Efficient 3D Semantic Segmentation with Superpoint Transformer | |
| PointTransformer | 73.5 | Point Transformer | |
| PointTransformerV2 | - | Point Transformer V2: Grouped Vector Attention and Partition-based Pooling |
0 of 6 row(s) selected.