Technology

Instead, TMOS researchers used metasurface-based upconversion technology, which essentially provides an easier pathway for light photons to be processed. The photons travel through a resonant metasurface, where they mingle with a pump beam. The non-local lithium niobate metasurface boosts the energy of the photons, and draws them into the visible light spectrum without the need to convert them to electrons first. It also doesn't require cryogenic cooling – which reduces 'noise' for sharper images in traditional night vision – so can do away with even more of the bulky night-vision goggle mechanics.

This reads like a Turbo Encabulator script.

This isn't really night vision in the typical sense. It's an Infrared camera in a thin package.

Also Military night vision is described wrong. The photon doubled is quite small. The problem is that afterwards the image needs to be turned again. That is done with fiberoptics. Those take the amount of space.

Can't wait to add this to my transitions blue light filter colour blind prescription smart glasses.

First: I'll believe it when I see it. Every so often pie-in-the-sky claims of this type come out, and they often end up not being feasible, even if they're technically possible.

Second: if it is feasible, given that gen 3 night vision tubes have remained stubbornly expensive, I would not expect this to be cheap for a long time.

I don't have much to contribute to the technical discussion here, just my comment that even playing with kids toy night vision goggles is awesome. For about $100 you can buy a really fun toy to play around with. Gets boring quickly, but kids might have fun with it longer.