What specific keywords do I need to create the following in generative AI

[Matthew249224202mcs]

I'm finding that creating anything "real" is where Generative AI has the most problems. 

For example, I want to create a background of a baseball diamond, looking at first base from about 10 feet away. I want it to be real, not illustrated or animated or surreal or painted, and I want it to not have people in it. 

So far, Generative AI  hasn't been able to produce a reasonable facsimile of a photorealistic empty baseball diamond, let alone any of the other requirements.

What I end up with is usually some impressionist rendition of a cricket pitch, with a sidewalk in the middle... 

Is there a dictonary of practical keywords to use?
Is Generative AI capable of doing what I've described?
Why is Generative AI so unintuitive compared to any other AI platform I've experimented with? If it's my way of thinking that needs to be updated, that's fine...

Thanks in advance.

Matt

Thanks!

[Kevin Stohlmeyer · Accepted solution]

Instead of Generative Fill - have you tried Adobe Firefly instead? You can choose Photo instead of art and generate completely new art.

This page gives help with generative prompts:

https://helpx.adobe.com/firefly/using/tips-and-tricks.html

"baseball diamond view from first base ground view"

baseball diamond view from standing at first base

You have to realise that Firefly and Gen AI were release just this past summer, so not even a year old compared to other offerings that trained on larger (albeit illegal) sources, and have had a longer development period.

[Matthew249224202mcs]

you're my big damn hero 😉 

Thanks.