Class CorrectPointCloudMotion
Represents a transcoder taking an input point cloud distorted by motion, and generating the corrected output point cloud.
Inherited Members
Namespace: Mechatronics.SensorSDK
Assembly: solution.dll
Syntax
public class CorrectPointCloudMotion : TranscoderRemarks
If a lidar moves during acquisition, each depth is measured at a different moment in time, and thus, each point has a different reference frame. This class adjusts the points so they are all expressed in the last pose of the reference frame. The correction assumes temporally equidistant points. If using a lidar with a custom firing sequence or a non-360 field-of-view, the correction will be inaccurate.
Fields
| Name | Description |
|---|---|
| _correctedPointCloud | Point cloud where motion distortion has been corrected. |
| _prevReferenceToWorldPose | Reference frame pose during the last call to Transcode(CustomPassContext). |
| enableCorrection | |
| inPointCloud | |
| outPointCloud | |
| referenceToWorldTransform | The Transform in which the point clouds are expressed. |
Properties
| Name | Description |
|---|---|
| ReferenceToWorldPose | Retrieves the current pose of the reference frame. |
Methods
| Name | Description |
|---|---|
| Disable() | Disable change event on the inTranscode port and release the graphics resources. |
| Enable(ClockState) | Initialize the transcoder shader and enable on change event on the on inTranscode port. |
| Transcode(CustomPassContext) | Add to the command buffer the conversion of the point cloud buffer into a depth buffer. |