Has everyone seen this paper from 2016?
I'm really curious about thoughts regarding this, particularly the performance of the ML method, from the upper echelons...
Dual iso has so much promise. It requires a new approach to achieve any of its potential.
https://repositori.upf.edu/bitstream/handle/10230/28210/Bertalmio_ElectronicImaging2016_high.pdf?sequence=1&isAllowed=y
does no one have any thoughts about his approach?
They describe the algorithm, I didn't spend much time on it but it looks like there's enough info to implement it. If someone was willing to spend the time to do that, it would be possible to compare the methods.
What kind of discussion were you hoping for?
I'm curious if anyone who is familiar with the current approach would have any thoughts or inspiration resulting from reading it.
Okay. Mine is basically: sounds like it would be interesting to test it.
I wrote to one of the authors and he said that the solution was quite inefficient and they sadly didn't develop it any further, so no code available.
I guess it could have worked for single images at least.
That's a shame, academics will nearly always share their code. At least you tried :)
There's still enough info in the paper to attempt to use the technique, though obviously that's more work.