Results for SpecOT
Submission data
| Full name | Spectral Optimal Transport for Unsupervised 3D Semantic Segmentation |
| Description | SpecOT is an unsupervised 3D semantic segmentation method that performs spectral primitive discovery through risk-aware optimal transport, enabling reliable pseudo-label generation from unlabeled point clouds. |
| Input Data Types | Uses Color,Uses Geometry Uses 3D |
| Programming language(s) | python with cuda |
| Hardware | GeForce RTX 4080 |
| Submission creation date | 15 Jun, 2026 |
| Last edited | 15 Jun, 2026 |
| Last uploaded | 30 Jun, 2026 |
3D semantic label results
| Info | avg iou | bathtub | bed | bookshelf | cabinet | chair | counter | curtain | desk | door | floor | otherfurniture | picture | refrigerator | shower curtain | sink | sofa | table | toilet | wall | window |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.398 | 0.537 | 0.635 | 0.331 | 0.184 | 0.617 | 0.288 | 0.436 | 0.346 | 0.300 | 0.894 | 0.105 | 0.000 | 0.014 | 0.612 | 0.173 | 0.544 | 0.355 | 0.469 | 0.651 | 0.464 |