Results for PBNet
Submission data
Full name | Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization |
Description | Binarization and classification of each point in the point cloud |
Input Data Types | Uses Color,Uses Geometry Uses 3D |
Programming language(s) | C++ Cuda python pytorch |
Hardware | 3090 |
Submission creation date | 25 Nov, 2021 |
Last edited | 9 Mar, 2023 |
3D semantic instance results
Info | avg ap 50% | bathtub | bed | bookshelf | cabinet | chair | counter | curtain | desk | door | otherfurniture | picture | refrigerator | shower curtain | sink | sofa | table | toilet | window |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
![]() | 0.747 | 1.000 | 0.818 | 0.837 | 0.713 | 0.844 | 0.457 | 0.647 | 0.711 | 0.614 | 0.617 | 0.657 | 0.650 | 1.000 | 0.692 | 0.822 | 0.765 | 1.000 | 0.595 |