… is a picture I took this morning while I sat with a cup of coffee.
If you have been following along with these posts recently, you will have noticed that a lot of images I use in the 366 series are generated by an AI. (More often than not Leonardo.ai) I do not spend a lot of time on creating the perfect prompt. I am really just playing around and creating a little eye candy whilst learning what I can. Which got me to wondering, what would Leonardo make of my view?
I used this prompt.
“I am looking at Auckland Harbour from a balcony in the viaduct. The sun is shining, but the day is still quiet. Create an image for me that you think reflects what I am seeing.
I ran the prompt three times and these were the results.
I decided this was the closest to the right result as in ‘matching the prompt’.
- It has Sky Tower - so ‘Auckland’ - except if I was looking at the harbour from a balcony - Sky tower would be behind me.
- It is very clearly a balcony.
- I rejected the bottom left, because really - to match the prompt ‘I’ wouldn’t be in the picture.
- There were others that were as acceptable. None were ‘right’/
So - why no boats in the AI images? Because they weren’t included in the original prompt.
Just for kicks I modified the prompt
“I am looking at Auckland Harbour with some moored boats from a balcony in the viaduct. The sun is shining, but the day is still quiet. Create an image for me that you think reflects what I am seeing.”
Am I happy with the finished results? Well - no.
But if i didn’t have access to the actual image, for a few minutes of work - the results weren’t too bad.
(These are the results from the free version of Leonardo. I know they would be massively improved if I subscribed and added access to the other models they have available. First though, I want to learn as much as I can for free.)
What do you think?
