I do a lot of amateur astrophotography too (https://www.instagram.com/dheeranet/), mainly with the objective of trying to show what things "would look like" if these deep space objects were much brighter; I maintain compositional and visual accuracy; all of these things could have theoretically been shot as-is as a single image if the objects were brighter.

> It's not uncommon to combine the LRGB data with e.g. H-alpha data.

Yeah I do this too. Typically I do HOO to maintain as close to spectral accuracy as possible. I don't have an S filter.

It is, by the way, possible to get DSOs shot as a single image, it's tricky. See the 3rd image in this series:

https://www.instagram.com/p/CKXdL1YndQi/?img_index=3

I want to try to fine tune a diffusion model to make it possible to actually shoot these single shot with good fidelity.