I have a dataset containing 3D point-cloud image data, ie I have four coordinates, x,y,z and Intensity. I am tasked with coming up with a generative model that produces new images based on the dataset.
It's not very clear how I would feed my model data of this kind. I know that I can flatten an image into an array and feed it in if I had a grid-type situation but I don't know how to translate that in the case of a point-cloud. Any suggestions are welcome.