• RegalPotoo@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    2 months ago

    The most unbelievable thing about these photos are that trump could kneel like that without shattering

    • nonfuinoncuro@lemm.ee
      link
      fedilink
      arrow-up
      0
      ·
      2 months ago

      except for the made up CHINESE characters behind the north KOREAN man.

      !as an aside historically Korea Japan and Vietnam actually did use Chinese characters and still do so for various purposes (Japan the most) but nk explicitly outlawed their use for nationalism reasons !<

  • Mossy Feathers (They/Them)@pawb.social
    link
    fedilink
    arrow-up
    0
    ·
    2 months ago

    The awkward kiss almost convinced it was real. Almost. The kneeling photo is a bit jank and hard to believe, but the kiss being so awkward is so believable that it almost convinced me lol.

  • werefreeatlast@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    2 months ago

    Wow! No decency! trump should have at least tried to suckle on his supreme leader. He loves everyone who is supreme like that bigly.

  • collapse_already@lemmy.ml
    link
    fedilink
    English
    arrow-up
    0
    ·
    2 months ago

    The photo of him saluting the N. Korean general is real. We don’t even need to fabricate this kind of material. If we are going to fabricate it, Trump’s service should be more explicit and x-rated.

    • Asafum@feddit.nl
      link
      fedilink
      arrow-up
      0
      ·
      2 months ago

      I like the idea of fabricating it specifically to use on Twitter since Elon is being a piece of shit about it all.

  • tfw_no_toiletpaper@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    2 months ago

    AI has come to the point where I only have a “gut feeling” it is AI because it looks so polished and weird. But there are many aotzer (stolen) art-styles that look completely different and I wouldn’t notice if no one told me.

    you cannot sniff them out anymore by just saying “look at the wretched fingers”

    • imPastaSyndrome@lemm.ee
      link
      fedilink
      arrow-up
      0
      ·
      edit-2
      2 months ago

      you cannot sniff them out anymore by just saying “look at the wretched fingers”

      Really? Not even the totally detached hotdog on the general in the back?

      Him playing air guitar. Missing a finger on the other side

    • Venator@lemmy.nz
      link
      fedilink
      arrow-up
      0
      ·
      2 months ago

      The lighting is always way too good, looks like a model shoot or 3d animated movie where a team of people spent ages tweaking all the light positions and brightness.

  • RememberTheApollo_@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    2 months ago

    Which AI does realistic representations of real people? I’ve tried some and they can never get the likeness right. Or it seems they might get the face close, but the scene you’re trying to create wrong.

    • fross@lemmy.world
      link
      fedilink
      arrow-up
      0
      ·
      2 months ago

      Look on communities for stable diffusion or flux. The latest stuff is eerily good.

    • michaelmrose@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      2 months ago

      I asked chatGPT to draw Trump with Biden as a shark and instead of refusing it literally drew Hitler in a boat being threatened by a shark in Aviators

      • Zron@lemmy.world
        link
        fedilink
        arrow-up
        0
        ·
        2 months ago

        Looks more like 2 lovers who found out a close friend died.

        I’ve seen my cousin look like this with their spouses at funerals.

    • linearchaos@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      2 months ago

      AI isn’t that magical. You get good images by continuing to feed it prompts and tweaking weights until it eventually makes something good. Lighting kinda suggests mid journey, but it could as easily be 1000 tries of stable diffusion.

      • RememberTheApollo_@lemmy.world
        link
        fedilink
        arrow-up
        0
        ·
        2 months ago

        I’m quite familiar with AI and how the prompts work. I’ve been using a couple for years. Each one seems to have its own particular (Skill? Emphasis? Personality?) and no matter what you feed it via prompts it may not overcome whatever it is that prevents it from generating the scene you want. That may depend on what it was trained on - say for instance you want a character from Labyrinth (1986) but instead the AI keeps forcing imagery from Pan’s Labyrinth and no amount of negative or other prompts will fix it. Some are deliberately designed to avoid things like accurate likenesses (some devs don’t want their AI used for deepfakes), some emphasize an art style, and some like StableD require tons of “corralling” to get it to generate what you want - if you’re lucky.

      • olicvb@lemmy.ca
        link
        fedilink
        English
        arrow-up
        0
        ·
        2 months ago

        @[email protected]

        Also if it’s on Stable Diffusion there’s 100% a LoRA for those two available. So from that point it would be harder not to get their likeness :D

        • RememberTheApollo_@lemmy.world
          link
          fedilink
          arrow-up
          0
          ·
          2 months ago

          I’ve not used StableD directly to create a specific person, or at least not popular/known ones, so haven’t had the benefit of a LoRA. I generally just do scenes that, if they have people, they’re just background fillers or any likeness will do.