Github: https://github.com/shobrook/densify

How does it make assumptions about the depth of those 100 points it just makes up from the 3?

Nice work! How did you create this video?

I feel like deep learning is overengineering this problem

You're correct, but that's not what he's doing.

This is a tool with deep learning applications, not a tool that uses deep learning.

I used FuncAnimation in matplotlib. I made it a feature of this package to visualize the process of "densifying" a given point cloud.

Not sure I understand what you mean by depth.

How do you choose which points to connect?

I see, was a bit confused from the title of what your goal was

Isn't this just Voronoi interpolation?

You seem to be asking for the purposes of photogrammetry right?

I presume it could be scaled to 3D easily from 2D

But also I don’t know much and am stupid