Convert to Dense Point Cloud

The algorithm converts a given Point Cloud to a “Dense Point Cloud”, i.e. a depth or range image of the point cloud from a certain point of view. This is mainly intended to re-add lost point-to-pixel mappings for data that already is “dense” and for which the user has a matching view matrix and camera intrinsics.

Input

The algorithm takes exactly one Point Cloud.

Output

The input object will be modified in-place by replacing position, color, and normal data with the transformed data (via the view matrix) and adding the point-to-pixel mapping.

Description

The user needs to set the View matrix, Camera intrinsics matrix, and image resolution (Viewport width times Viewport height), which needs to be given by some outside information source (usually provided with the data).