RE: Can "AI" really improve a quality photographic image?
You are viewing a single comment's thread:
That is a good shot. I don't think the AI helped though I did enjoy the "art" one XD
Sometimes while these things are marketed towards professionals to enhance their workflow I feel like they're actually targeted for normal people to make them feel like they're producing good art without having to actually produce good art.
Ages ago sibling dearest shared an article with me about someone claiming artists were just butthurt because AI was "levelling the playing field" and that they were "looking forward to developing their talent" (from vague memory they didn't do any additional work on the images they generated). And I remember reading on Reddit where AI proponents were very confidently asserting that soon people would have "no choice but to accept they were artists too".
And that's usually all I have stuck in my head despite knowing that there are people out there using it for photobashing or to knock out super fast concepts for later refinement or to clean up lineart and drop flats and whatever else you can do with it as another tool.
J agrees with your sentiment (he doesn't like editing photos after). Meanwhile I have historically gone for fast over properly set up because kids move fast and I prefer candids to posed even though we also had a lot of fun doing "photoshoots" XD so if I didn't have one good shot out of the fifty million or so I would take then the best out of the bunch would get some editing to try to salvage it x_x
I'm too much of a "romantic" for taking things too far from the truth in photography. Roland Barthes "Camera Lucida" is about 80% of my belief in photography as a powerful, moving medium. Come to think of it, its been something like six months since I last read it, maybe it needs another cycle through my brain...
Sure, there's the one click (and VERY expensive tools) that get you pro-level, frequency separated, dodged, burned, textured, skin, but... that's not the objective reality we find our selves in.
I am in a bit of a difficult position emotionally about the AI Art / LLM side of things.
I hate LLMs right now because people don't understand how they work and use them to replace thought. I like text to image, because however the image is generated, it makes me force myself to think and look for elements that I wouldn't envision in that image if I was to draw it from my own imagination, a sort of regressive analysis on "I wanted this, but this element is there for a reason, why?"
I don't know. I have had the classical art training. I know how to "Speak" art. So many people who are "prompt engineers" or "prompt seers" are just hitting the right, ticklish spots of that black box to get "nice" images.
But when people with proper art training start using it, and start photobashing, and start making, I think that's when there's the possibility for magic. The possibility, because not everything every artist produces is a hit every single time.