3D point cloud is a commonly used data in spatial modeling & understanding. Such a data can be captured from laser scanners, stereo camera, or structured-light cameras. Point cloud processing is essential for applications like self-driving cars, infrastructure inspection, construction modeling, etc..

Here we will go through some basic modules of a point cloud processing pipeline, include 1) clustering; 2) normal vector estimation; and 3) triangulation.

1. Week 1: Point Cloud Basics.
2. Week 2: Proximity and Connected Components.
3. Week 3: Local Surface Normal Estimation.
4. Week 4: Clustering via Proximity and Local Normal Alignment.
5. Week 5: A Simple Triangulation Scheme.
6. Week 6: A Simple Approach for Point Clouds to Surfaces.
7. Week 7: Bridge Visualization and Comparison. Do Part #1.
8. Week 8: Bridge Comparison. Do Part #2.

Here we cover 1) point cloud registration and 2) point cloud descriptor extraction.

1. Week 1: Aligning matching point clouds
2. Week 2: A Basic point cloud descriptor
3. Week 3: Local descriptors.

Week #X: Point Cloud Segmentation

/* Go through the page on region growing. As you may notice, there is a sample code of region growing segmentation using PCL on that page. Simple copy and run that code won't count.

For Matlab folks, an exemplar implementation of region growing can be found here. Note it's implemented for 2D image, and you need to extend that to 3D point cloud.

For PCL folks, you may refer to the sample code. Try to implement your own version of region growing. */

/*__** Week #X: Cloud Proximity to Another Cloud

Get set of points that a given point cloud is near to relative to another point cloud.

Notes: Matlab function: pdist2 */

ECE4580 Learning Modules