• @vivavideri@lemmy.world
    link
    fedilink
    428 months ago

    Okay but As a native north carolinian and a csi enthustiast, my only true beef with this and any other cinematic representation is… NC doesn’t require front plates. Just the rear. So when people filming over in cali are all like, yo, pop this NC plate on the front bumper, I just… i just lose it. From a continuity standpoint.

    The csi zoom screw passes the vibe check for me, but the NC front plate does not lol

  • Mossy Feathers (They/Them)
    link
    fedilink
    248 months ago

    There’s gonna be a day when people don’t understand this anymore. Considering the resolution phone cameras have now, it may not even be that much longer.

    • @DrugsMcChrist@lemmy.ml
      link
      fedilink
      148 months ago

      I’m wondering if AI can already solve this. I’m not even some crazy AI fanboy, I’m just thinking about the possibility of predictive AI being able to interpret compression artifacts to determine what forms would collapse into a particular pattern.

      • @AeroLemming@lemm.ee
        link
        fedilink
        English
        348 months ago

        It’s physically impossible to perform any type of analysis on a low resolution video and end up with a massively* upscaled and accurate result no matter what sort of technology you use. AI just guesses from what it can see and what it’s been trained on. It’s the digital equivalent of hiring an artist to create a higher quality rendition of a picture.

        *There are some tricks that can be used to “enhance” images and especially videos, but not so much that it’s like having optical zoom. For example, a bunch of noisy video frames of the same thing can be combined to get a less noisy result.

      • @hark@lemmy.world
        link
        fedilink
        238 months ago

        No, but AI can be used to make up “evidence” and falsely convict people through “science”, “technology”, and “math” that people don’t understand but assume is correct because computers or something.

      • @Knusper@feddit.de
        link
        fedilink
        148 months ago

        There have been upscaling AIs for a few years, which can take a blurry picture and then e.g. guess that some pixels are probably hair, so it’ll swap those out for a custom rendered version of hair.

        Sometimes that works well, but you often still have Uncanny Valley stuff going on. I also certainly don’t feel like they’re better at actually interpreting low-res images than humans, not in their current state.

        And well, it should also be noted that if you prime such an AI with an image of the suspect, it will absolutely find a way to make a blurry mess of pixels look like that. So, it certainly shouldn’t serve as the only evidence.

      • HonkyTonkWoman
        link
        fedilink
        48 months ago

        Calm down there, you’re starting to sound all inquisitive & such. Like that creamsicle lookin’ fella from the show.

  • @HiddenLayer5@lemmy.ml
    link
    fedilink
    English
    118 months ago

    Honestly seems that a lot of crime shows would become more realistic if they just had a psychic on the squad.

    • @anton
      link
      58 months ago

      Answer: A ML-model just makes it the fuck up