hdr-plus-swift

For those of you who are an apple, someone implemented Google's HDR+ to work on raw files from DSLR cameras: https://github.com/martin-marek/hdr-plus-swift

It still seems we're talking about completely different things, or maybe you're trying to derail the conversation... Both ML-based processing and tonemapping take place after demosaicing to produce the final JPEG. RAW capture is just your typical laplacian pyramid stacking/exposure fusion with motion compensation and sensor-specific calibration. That's the main contributing factor that somewhat compensates for the sensor size, not some kind of ML processing. If you think that ML is somehow involved here, I have zero idea where you are getting it from. The entire point of pre-demosaicing pipeline is to robustly converge the captured stack into a neutral scene-referred baseline with provably better SNR and precise photometric parameters. And that's what you get in a resulting DNG, not some neural tonemapped RGB image (which is not even crea

> By low light mode on phones I assume you mean image stacking. This requires a stationary subject. Which is a misconception. The entire point of stacking in smartphones is to handle lots of movement in a robust way, and it's surprisingly good at that. You can actually perform similar stacking manually for a camera, the algorithms are well known and there are applications to do just that, for example [1]. There's less benefit on a normal-sized sensor since it can perform much better with a single shot as you pointed out, but there are still valid uses for it in certain scenes where even proper cameras struggle, see [2]. [1] https://github.com/martin-marek/hdr-plus-swift [2] https://www.timothybrooks.com/tech/hdr-plus/