Point Cloud Preprocessing

The main function of this step is to generate the point cloud, remove noises and then obtain the point cloud in the ROI. It usually follows the step Capture Images from Camera to accelerate. The combination is as Figure 1. shown.

Figure 1. Procedure of point cloud preprocessing

The introduction of each steps is as follows:

1. From Depth Map to Point Cloud Input the original colored image and its depth image, and output the point cloud of the current scene.

2. Calc Normals of Point Cloud and Filter It Calculate the normal vectors and remove useless points of the point cloud obtained, amd output the filtered point cloud with normal vectors.

3. Extract 3D Points in 3D ROI Input the processed point cloud and then extract part of the point cloud in the ROI.