Results for PTv3-PPT-ALC
Submission data
| Full name | Point Transformer joint training with ARKitScenes LabelMaker Consensus Dataset |
| Description | We use LabelMaker to automatically annotate ARKitScenes, one of the largest real-world indoor 3D scene dataset. We add this dataset to PTv3-PPT joint training. |
| Input Data Types | Uses Color Uses 3D |
| Programming language(s) | Python with CUDA |
| Hardware | 8x RTX 3090 |
| Submission creation date | 5 Mar, 2024 |
| Last edited | 25 Oct, 2024 |
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
 | 0.798 | 0.911 | 0.812 | 0.854 | 0.770 | 0.856 | 0.555 | 0.943 | 0.660 | 0.735 | 0.979 | 0.606 | 0.492 | 0.792 | 0.934 | 0.841 | 0.819 | 0.716 | 0.947 | 0.906 | 0.822 |