Hi, newbie here. And as a newbie, I retain the right to ask an obnoxious and overly ambitious question on how to achieve an effect that no one is interested in achieving 
In brief, I’m trying to ‘glitch’ render a house so that it’s turned inside out. More specifically, I would like to set up the camera with a view of the house’s exterior, but only render interior faces/objects that normally would be occluded behind the exterior walls. And to complicate things more, is there a way so that what gets rendered is based on pixel depth sorting?
After pouring through various threads and posted questions, I think my question posed in more technical terms would be:
Is there a way to customize depth sorting so that based on an object’s hierarchical Z depth at each pixel location, I can choose which order of pixel depth to render? For example, I only want to render the second closest pixel depth relative to the camera. In my conception, this is how the image will become glitchy as you change perspectives.
I hope this is clear.
Theoretically, can this be done in blueprint visual scripting, at the coding level, or would require accessing the gpu?