That, but I also think it's a really hard, abstract thing to train the models on regardless.

I could be wrong about this! Maybe it's easier than I think. But it's not like you can just say to the model "oh yeah, and make sure all the edges of things follow the rules of perspective." It has to learn those rules the same way it learns everything else - basically, by looking at a bunch of examples and getting a "feel" for what's right. (Well, "a feel" = "the values of the model's weights updated to produce this result" and so forth, but yunno.)

But it's not the kind of detail that immediately jumps out, as long as it's not *too* wrong. Observing it requires both figuring out which lines are relevant, and knowing how those lines should behave, and image-gen AI has no special ability to do either of those things. It has no ability to follow rules precisely.

The fact that human brains can also look at the pictures and not immediately go "wait, that's wrong" gives me confidence that AI models won't get it either. Even humans generally need to get out a ruler and start measuring. I think it's hard for human brains to just see it for pretty much the same reason it's hard for AI, but until AGI is a thing, strategies like "know the rules concretely" and "draw a line with a ruler" are more or less out of reach for the AI.

Oui, c'est faisable.
Mais se concentrer sur X c'est délaisser Y.
Au début, quand il fallait compter les doigts des mains, les modeles qui étaient bons sur les mains étaient mauvais sur le reste.

L'amélioration n'est venue qu'en multipllant le nb de paramètre des modèles. Et donc le coût de génération d'une seule image.

C'est exponentiel.

Et j'ose croire qu'il n'y a plu moyen de multiplier encore par 2 leurs coûts, sans revenus.

Générer plein d'images par IA, demander a des petites sous payées de dessiner les lignes fuites. On fait deux jeux de données : les images avec un seul point d'intersection et les autres. On rajoute des vrais images dans la première catégorie. On lance l'entraînement d’un modèle ou un fine tunning d’un modèle existant.

@FabMusacchio

It's not the sun's rays that meet at a point, it's the lines from the objects' shadows to the corresponding points on the objects that should meet at a point.

The statement about the sun's rays being effectively parallel just means that the direction of the light source can be considered the same for all objects.

Put simply, the consumers of the AI generated images do not care whether or not all the lines properly converge onto a vanishing point. Human vision may care about weird extra fingers, but vanishing point convergence? Nope. Don't care.

Human viewers will never notice these perspective errors, so AI models have no incentive to fix them.

Good points… except the bad one: the dinosaur graphic shows a line connecting different toes to the horizon

nothing makes sens. the chains ?? so much to say about them. Why ? where are they attached ? Why are they so unconsistent ?

The uniforms that are not uniform. like. You had one job. The helmets… the nonesense "text". the collars. the trousers too large ? In the military ? I don't buy it. the shirt that looks heavy and "cargo" at the top bit they look light like an everyday summer shirt painted green. And what are those white dots ? The masks, what... what are are they ? are they masks or chinstrap ? and what do they seam to eat them, are they nervous ? They you look at the figures behind and find that they make less and less sens at each row. they sort of melt.
And The guns. what are they. you can't even identify each parts of the weapon. and then you look a the hands. why are they blended ?

and. look between the legs of the front figure. it like there are no one behind him.

I could go on all day about so many obvious mistakes.

It feels like an "early" Dall•e picture the way it kind of seems painted and blended.

As a photographer who is very fond of extremely wide angle lenses, I can tell you these tests will often fail on real images, as well. But the points made are valid...