Submitted by Manuel Dahnert.
Submission data
| Full name | Fast Point Feature Histogram + 3-Point Ransac |
| Description | We use the FPFH (Rusu et al.) as local feature descriptor to compute correspondences between Haris keypoints of the CADs and of the scan.
We filter the correspondences and compute a transformation using a 3-Point Ransac. |
| Input Data Types | Uses Geometry Uses 3D |
| Programming language(s) | C++, PCL |
| Hardware | Intel Xeon E5-1620 v3 |
| Submission creation date | 21 Jan, 2019 |
| Last edited | 22 Jan, 2019 |
CAD Alignment
| Info | avg (per CAD) | avg (per class) | bathtub | bookshelf | cabinet | chair | display | other | sofa | table | trashbin |
| 0.037 | 0.020 | 0.000 | 0.000 | 0.015 | 0.107 | 0.000 | 0.012 | 0.021 | 0.029 | 0.000 |