Results for ALS-MinkowskiNet
Submission data
Full name | MinkowskiNet pretrained with ARKitScenes automatic labeled dataset |
Description | !!! IMPORTANT !!!
The short and full submission name is misleading, because I cannot contact the benchmark maintainer and hence re-use the submission slot. This submission uses PTv3-PPT model. 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. |
Publication title | ARKit LabelMaker: A New Scale for Indoor 3D Scene Understanding |
Publication authors | Guangda Ji, Silvan Weder, Francis Engelmann, Marc Pollefeys, Hermann Blum |
Publication venue | CVPR 2025 |
Publication URL | https://openaccess.thecvf.com/content/CVPR2025/html/Ji_ARKit_LabelMaker_A_New_Scale_for_Indoor_3D_Scene_Understanding_CVPR_2025_paper.html |
Input Data Types | Uses Color Uses 3D |
Programming language(s) | Python with CUDA |
Hardware | 8x RTX 3090 |
Website | https://labelmaker.org |
Source code or download URL | https://github.com/quantaji/LabelMaker-Pointcept |
Submission creation date | 22 Feb, 2024 |
Last edited | 4 Jul, 2025 |
Last uploaded | 13 Oct, 2024 |