I'm saying it because it's not only obvious with even a moments thought (you can literally just ask it for an entirely red image or whatever), but also because it's easily provable.

Prompt: "Under the sea"

Image:

Average pixel colour:

Prompt: "a man with red hair wearing a red coat standing in front of a red background"

Image:

Average pixel colour:

So I ask you the same question. Did you just say that because you felt like it was true?

[–] Krauerking@lemy.lol 0 points 4 days ago (1 children)

The average brightness values of those are both middle of the road grey. Sorry I should have rephrased as I misspoke calling it beige but the point still stands that has the most average toned color.

If you look they are middling around 50-60% where as a similar red photo intake would likely have a higher contrast and an average color with a higher brightness.

[–] vithigar@lemmy.ca 2 points 4 days ago (1 children)

I'd expect that many images are going to be somewhere near 50% grey if you average their luminance out overall. That's just the average of every colour though. The fact that averaging a range of things tends toward a standard distribution isn't particularly surprising. Again though, it's not hard to get a diffusion model to generate something outside of that expectation.

Prompt: "night sky"

Image:

Average colour:

Average brightness: 21%

Prompt: "lineless image of an old man drawn in yellow ink on white background"

Image:

Average colour:

Average brightness: 90%

[–] Krauerking@lemy.lol 1 points 4 days ago* (last edited 4 days ago) (2 children)

Ehhh you'd be surprised how much small highlights and true dark values will skew an average. You don't really get blown out or completely under exposed parts of AI generated images. It keeps trying to add some saturation.

I will say I'm getting different brightness levels than you are but it's not a big deal. Point made.

However I will say saturation is also pretty much 50% across the board for all the colors so maybe there is something there to use as an indicator. The average color always comes out grey toned somehow.

Testing it myself has been pretty spot on.

Just to point it out, here is a photo I took to mimic the red man. And the color average from it.

It doesn't even have the highlights of the face and yet it's already much brighter and has a much higher saturation.

[–] vithigar@lemmy.ca 2 points 4 days ago

Also, this is tangential to the rest of our conversation, but I appreciate the dedication to the comment chain required to actually set up something with similar composition to the red man image and take a picture of it. Even has some black in the image in roughly the same size and area as his sweater. :D

[–] vithigar@lemmy.ca 1 points 4 days ago (1 children)

I'm not sure what you mean by the saturation being around 50% across the board. If I peek the HSB of all of the averages only that first teal-ish one appears to be around the mid point for saturation.

[–] Krauerking@lemy.lol 1 points 4 days ago (1 children)

When I look at the HSL of those colors I get 60, 54 and 46 (ignoring the ocean one cause I just don't have that open anymore) for the S (saturation) value.

Like I said it's weird that we are getting different values cause your brightness also wasn't in line with what I had. And that wouldn't be a screen issue.

My red came in at 73% for me.

[–] vithigar@lemmy.ca 1 points 4 days ago* (last edited 4 days ago) (1 children)

Odd. I tag your red at 78%. And for what it's worth this RGB to HSV converter agrees with that number taking your colour hex as C92D20. I certainly don't know enough about it to offer an explanation as to why it might be different.

edit: Ah, I think it's HSV vs HSL, which I'm just now learning are different things. :D

[–] Krauerking@lemy.lol 1 points 4 days ago (1 children)

I am importing mine into a photo editor so might be something with that. Anyways.

Thanks for the examples. It gave me something to think about and the discussion has been interesting. I think there may be details to take away from this but I'm not in a rush to dip back into AI images.

[–] vithigar@lemmy.ca 1 points 4 days ago

For what it's worth I agree that AI images will generally have "tells" that give away their nature. It's just they aren't quite so straightforward as being able to check that average values are within a range. It would be nice if it were that easy though.

While I do dabble with AI image generation I'm not a lunatic who calls themself an "artist" for doing so, nor do I think being a "prompt engineer" is any kind of expression of creativity or skill. I think the people who do are completely self-deluded.