Probably not much harder, but you wouldn't get the same massive jump in quality that you get going from 1 image to 2. NeRF/Gaussian Splatting in general is what you're describing, but from the looks of it, this just does it in a single forward pass rather than optimising the gaussian/network weights.

That’s already how Gaussian splats work.

They’re novelty of splattr (though I contest that they’re the first to do so) is that they need fewer images than usual.

I think the novelty is that they don't have to optimise the splats at all, they're directly predicted in a single forward pass.

That’s not really novel either imho, though google search is escaping me on the specific papers I saw at siggraph.

Imho it’s an interesting combination of technologies but not novel in an off itself.

The novelty here is that it does work on uncalibrated images.

A lot of splats systems do work on uncalibrated images so that’s not novel either. They all just do a camera solve, which arguable isn’t terrible for a stereo pair with low divergence.

Not really, it is using Mast3r to determine camera poses.