I use several raycasts from the specific points from the textures in the world to the magic camera texture. All the raycasts must hit a tiny collider on the player's camera. These raycasts will be aligned when the "forced perspective" is correct. It took a lot of tweaks, and I think it still needs more.
This was my first jam and I was short on time this week. Only got about 30 hours of work put into this, so I didn't have a chance to expand the number of puzzles. I had a much larger puzzle planned that used the same mechanic for other uses, such as moving platforms with the rune alignment (which is why the prototype level is very large). The deadline just came up too fast.
I was somewhat inspired by a combination of superliminal and Zelda Botw.
Also good to know it works with the Rift S. I only have an Index to test it on.