On 29/12/2020 10:16, Dr. A. Krebs wrote:
dt seems to require quite a bit studies to find out the appropriate steps to image
processing. As we can see, there are people searching for a beginners-type GUI.
For me some type of such "beginners workflow" helped a lot; however, there should
be a physical / arithmetical backup for such workflows.
Yes that makes more sense than a beginners-type GUI.
No matter the GUI, it is the ability to use it that really matters.
Overall, there might be some type of knowledge-system in the background to analyse
a selected pic. From there he system could offer a questionnaire to concrete the
targets and, at last, suggest some reasonable steps to improve a pic.
HO! What you are asking for is a AI like scene recognition and classification.
Yes they exist, online. Many exist. Some do things like 'sharpening' and colour
balancing and correcting skin colour. But they require more resources than you
have on your PC or laptop.
However yu DO have a tool that can do scene recognition etc etc and do the
correction the way you want not according to some algorithms in the AI determines
by a means even the original programmer wouldn't know.
It's called 'natural intelligence'.
It's called taking the time to learn tool.
If the idea proposed by some pundits that it will take 10,000 hours of practice to
fully master that craft puts you off, well the answer isn't "suck it up!'.
It's much simpler than that. It's like using a vacuum cleaner. You keep pushing
forward in the direction you care about bit by but, 'sucking up' little by little
as you go along. You don't need to master it all. What you do need to master you
don't need to master all in one go.
Any image is unique, and if you are anything like me, any film roll with have a
number of different types & style of image.
Dealing with every possible image in the way you ask is too much for this list.
It gets back to the AI situation.
All of the above ignores the fact that anything you do is not set in stone.
You can always re-edit the same image at a later date as your knowledge grows.
Even in one edit session you can experiment to learn. It's not like film and
paper where there is a delay before you see the result and cost to the material.
But if you DO experiment, don't rely on your memory. Have a plan and keep and
log. take a scientific, methodological approach. Logging failures and mistakes
is important too.
--
"Success is 20% skills and 80% strategy. You might know how to read, but
more importantly, what's your plan to read?" -- Jim Rohn.
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to [email protected]
