Convert 2D Points to 3D Points
Usage Scenario
This Step is suitable for scenarios where the quality of the point cloud actually acquired by the camera is poor. This Step combines 2D images (such as object masks detected by deep learning, object edges detected by 2D images, etc.) with their corresponding poses to generate the corresponding 3D point cloud data.
Input and Output
Parameter Description
| Parameter | Description |
|---|---|
Calc Normal |
Description: Once this parameter is selected, the normals of the output point cloud will be calculated. Default setting: Unselected |
Search Radius |
Description: This parameter is the neighborhood size used for calculating normals. Select Calc Normal to set this parameter. Default value: 10.0000 |
Iterations |
Description: This parameter is the maximum number of iterations for calculating normals. Select Calc Normal to set this parameter. Default value: 100 |
Error Tolerance |
Description: This parameter is the error tolerance of the RANSAC algorithm for calculating normals. If the value is set too small, too few points will be used to calculate normals, and if the value is set too large, more noise will be introduced. Both of the above conditions may lead to inaccuracy in normal calculation. Select Calc Normal to set this parameter. Default value: 1.0000 |