Semantic Segmentation on ScanNet200
43.5mIoUPLM
Evaluation results
| Method | Links | |
|---|---|---|
| PLM2025.09 | 43.5 | |
| ODIN-Swin-BInput Modality=Sensor RGBD Point Cloud, Backbone=Swin-B2024.01 | 40.5 | |
| ODIN-ResNet50Input Modality=Sensor RGBD Point Cloud, Backbone=ResNet502024.01 | 35.8 | |
| OctformerInput Modality=Mesh Sampled Point Cloud2024.01 | 32.6 | |
| CeCoInput Modality=Mesh Sampled Point Cloud2024.01 | 32 | |
| LGroundInput Modality=Mesh Sampled Point Cloud, Pre-training=2D CLIP2024.01 | 28.9 | |
| Diff2Scene2025.09 | 14.2 | |
| OpenScene-2DDimension=2D2025.09 | 12.7 | |
| OpenScene-2D/3DDimension=2D/3D2025.09 | 11.6 | |
| OpenMask3D2025.09 | 10.5 | |
| ConceptFusion2025.09 | 8.8 | |
| OpenScene-3DDimension=3D2025.09 | 6.3 |