I do as much research as I can. I take notes because I’m old school. First I see the idea in my head, but how to get there is not easy. I’m sharing because I guess that’s just who I am. Here are 2 prepared sketches done simply with magic markers.
That’s a lot of work for today. I’ll post this and then continue tomorrow for the next step. Please respect my art and do not steal. You are more than welcome to ask questions and I’ll reply. This method is something I create on a daily basis because I am trying to insert my personality into my AI creations. I don’t like copying from other people. Inspiration and copying are two different things.
When I look at women who are created in AI, so much of what I see is women who are 20 years old, picture perfect, either looking pretty plastic or they are all superwoman. I’m trying to see if I can create something a little different by trying a variety of verbal prompts to be more image specific and gear them to what people really look like, rather than having that “artificial quality” that I’m kind of sick of seeing. I am aware that models begin as photos of real people. That’s a given and there is a limited amound of changes one can make. People come in all shapes and ages and the models used often represent a kind of t.v. 1980s look. I’m not trying to create a super realistic look like a photo, but something that’s a combination of drawing, painting, illustration and storytelling without super hot colors and outrageous fantasy effects. Here is a little journey I did this morning exploring stages I went through. I used Leonardo AI and created the image with a simple prompt, using different styles, sizes and filters along the way. Basically the prompt I created was this. A woman architect of ….x age, with …..x colored hair, working in her office. She is professional and hard working, her clothes are faded blue denim, she wears glasses and the wall shows a mock up of one of her buildings.
I’m patching these in no preference, just as examples. I’m learning that I can make a woman look less plastic with using my own ideas in the word prompt. It’s a challenge and I have several other platforms I use and will experiment more with. It’s a time sink for sure and requires patience. Any likes and comments are appreciated. My idea in this project is that AI shouldn’t be hated and feared by artists. The more we learn, the more we grow.
I’m using Leonardo AI experimenting with crossing two separate images created independently and then mixing them up in the word prompt. I love the results of playing like this. The last picture is my original watercolor which is the source of the design. This is fun and it’s important to keep things super simple. This is not like MidJourney. This is simple illustration. It’s kind of like mixing apples and oranges and getting a weird smoothie.
This is my original watercolor in black and white:
Painting Mexico 
You must be logged in to post a comment.