Welcome to the 7th in a series of posts where I will test AI image generators and see how they handle making maze art. I will be asking 10 prompts and seeing what gets generated. My goal is to evaluate different AI image sites against each other to see how they perform. If you need to catch-up here is the series so far:
An exploration into Al Image Maze Generation
Making mazes with AI: Stable Diffusion
Making mazes with AI: Dream by Wombo
Making mazes with AI: Nightcafe
Making Maze Art with StarryAI
Today I test out StarryAI. You can access the website here. You must sign in to use the site and you receive 5 credits (and it seems 5 credits each day). Each prompt, or generation will generate 4 images and cost a credit (If you want 8 variations that will cost you 3 credits). There are 6 styles to choose from: No style, Fantasy, Portrait, 3D Art, Wooltize and Anime. You may also upload a photo to give the AI “something to start with”. All prompts I use are the same for each site !
Prompt 1 - Make a medium difficulty maze of the Eiffel Tower in black and white with arrows at the start and finish
I thought all 4 generations where interesting (normally there are some really bad ones). And I struggled picking between 2 of them. This one was a bit more interesting because the the perspective from above.
Prompt 2 - Draw a medium difficulty large maze of the Empire State Building with the start and goal embedded in the structure
Not really a maze, but it looks mazelike. 3 of the 4 images were really interesting and similar to this one. Made it hard to choose once again !
Prompt 3 - Draw a difficult maze of the White House pixel art style
All 4 images generated were different and none was close to what I was looking for. They did have a videogame like quality to them at least.
Prompt 4 - Draw a difficult maze that looks like a drawing of a famous building in sketch style
I liked all 4 images, I do not know what is ahead AI site-wise, but StarryAI is consistent in above average quality.
Prompt 5 - Draw a maze in the style of doyoumaze.com of a skyscraper in NYC
All 4 had this perspective looking up the side of the skyscraper. This one included a maze.
Prompt 6 - Draw a maze in the style of Sean C Jackson of a scene from a large outdoor market
It is a large outdoor market where they sell long white pathways ?
Prompt 7 - Make a maze of a slice of an orange in color
This is the first interpretation where the AI generated a maze like structure in the actual orange itself.
Prompt 8 - Make a maze integrated on top of a photograph of a king sitting on his throne looking cantankerous beside his beautiful queen
i must mention that 2 of the options included a queen, but I picked this one because the top of the chair has eyes, that king is dressed like Bond, James Bond with a fun look on his face, and the maze looks like brains. Just a lot to take in !
Prompt 9 - Make a solvable maze that is very large and very difficult to solve because it is so complex
All 4 were different and interesting. This is a solvable maze with many options if only we could read the sections on top and bottom.
Prompt 10 - Make a 3d render of a red and blue glossy cube maze
Here are 2 of the 4 generated. I like the one on the left a lot.
How did StarryAI do ? Consistent. I think there is a lot of potential here. I would mention again that I found it difficult to choose the best image multiple times. Easy to use. I recommend this site.
Coming Next: DeepAI