Using my own paintings as starter images with AI Art Generators

Regarding my traditional art doing abstract paintings with acrylic ink, watercolor ink or alcohol based markers has always been one of my favorite art activities. Now, that I do know a bit more about the possibilities of AI Art generators I wanted to use more of my traditional paintings. So I selected this painting, done with acrylic inks:

I did this painting in 2020. It is not one which I consider very well done, but I like the colors and the movement of the painting.

Now I tried out several Art Generators. I tried Blue Willow - which allows Image to Image generation, and I got a description of how to do it - but I somehow I failed again and again. I tried Dreamlike AI, which allows image to image generation and makes it easy to do it - but was not convinced of the result. Then I tried Lexica, and here is what I got:

As Lexica always generates four variations of an image, I had to select one of four. I must say I am pretty impressed with the result.

I also tried Deep Dream Generator with the same original painting. 

In my opinion the Deep Dream Generator version is more playful than the Lexica one. However I like both results, and will continue with combining my traditional art with AI Art.

The leisure time activities of Rocky Racoon

AI Art Generators make it possible to mix real subjects in a way to create images which are not real at all. I love racoons. In Germany racoons left their cages and went into the wild. Because of the fact that they have no natural enemies they prosper and even come into the big cities. I only know them from photos or from the one or other visit to the Zoo, but I love how they look. 

So I wanted to use AI Art Generators to create images of them - but not realistic ones showing them in their natural habitat. What about making them ride a bicycle?

So here is Rocky Racoon happily pedaling away on his bike. He even wears a helmet! I used the prompt:

Illustration of a cute racoon riding on a bicycle

As you can see I did not mention a helmet. Obviously AI connects "riding on a bicycle" with "wearing a helmet". 

But riding a bike is not the only leisure activity of Rocky Racoon. He also loves to read in his free time (Me too, by the way!) So I wrote the following prompt:

Illustration of a cute racoon reading a book

And voilà, here he is!

Looking at these images makes me want to write a story about this cute guy! But not yet...

Rose woodcarving

Woodcarvings are something I highly appreciate. In Germany and Austria an old tradition of carving nativity figures exist which are then sold internationally. There is also a German artist called "Tilman Riemenschneider" who was famous for his wooden carved out altarpieces. 

I paint, I draw - but I could never carve. I haven't got the eye for it, not to speak of eye-hand-coordination. But with AI Art Generators I am able to carve respectively generated images with carvings on them.

I decided to go for roses, and here is the first one, a wooden carving of a rose, created with the use of Nightcafestudio. 

The prompt I submitted was:

wood carving of a rose, highdefinition, hyperrealistic, intricate

Nice, isn't it?

However I decided to go to Lexica AI and submitted this prompt:

Wood carving of a beautiful rose, hyperrealistic, highdefinition, intricate

As you may notice, there is one additional word in the second prompt: "beautiful". Now have a look at the result: In my opinion the word "beautiful" makes the image richer, more detailed,

I must say, I prefer this image to the first one.

However I wanted to find out how Dreamlike AI handels the prompt. This time I used the very same prompt:

Wood carving of a beautiful rose, hyperrealistic, highdefinition, intricate

Dreamlike generated something slightly different, as you can see in the image below.

The wooden carving is not a realistic rendering of a rose, but very detailed and decorative. It could be an ornament on a piece of furniture.

Blue and White Chinese Ceramics

 You may ask: Why Chinese ceramics? Well, two things have led me to this subject:

• When I was a kid my mother had a friend who painted on porcelain, mostly flowers, butterflies and birds. I still have got some little vases and plates with her paintings on them. I remember watching her painting, being fascinated by her absolutely calm hand. I knew and still know that I could never do such a thing.
• I love history and I am very interested in the history of China. This does not only mean the social and political history but also the cultural and art history of this country, and here's where ceramics come in.

Of course I asked Chatgpt and was told that blue and white porcelain is one of the most famous types of Chinese ceramics. Knowing that I wanted to create such ceramics, focussing on vases and teapots (I love flowers, and I definitely love tea!). I wanted the vase and the teapot rendered in the same style. That's where Chatgpt came in for the second time. 

Here is the prompt for the vase:
Generate an image of an old Chinese teapot, hand-painted in cobalt blue with white and gold accents. The teapot should feature traditional motifs like dragons, phoenixes, lotus flowers, and intricate geometric patterns. It should appear aged with subtle cracks and signs of wear. The lighting should be soft, highlighting the vase's details. The background should be a neutral color. Aim for elegance and cultural heritage

This vase was created with the use of Nightcafe studio. As Nightcafe is limited in the number of words in the prompt, I had to shorten the prompt a bit, but it worked nevertheless.

Now I wanted to create a teapot in the same style. So I used the same prompt and replaced "vase" with "teapot". Actually I tried "plate" as well and got some nice results, but due to trying not to have a too long blogpost I will leave that out.

And here is the teapot:

The teapost was created with the use of Dreamlike AI, and as you can see, it is a bit more playful than the vase and obviously one would never dare to use it to actually brew tea in it. I would be afraid of breaking the handle!

Islamic Gardens

 A long time ago I had the opportunity to visit an Islamic Garden: There is a park in Berlin called "The Gardens of the World", and there you can see not only a typical Japanese Zen Garden, a Chinese Garden and a Korean Garden but also an Islamic Garden. I took a few pictures when I was there and just a few days ago I found these photos again. Among them was a picture showing a beautiful fountain.

Now, with my playing around with various AI Art generators I thought: Why not try to render a fountain in an Islamic Garden? So I opened Blue Willow and submitted the prompt:

"A beautiful Arabian fountain with flowers"

What I got was definitely more than a fountain. It was an Islamic Garden with the fountain or the water pool in the middle, as a focal point. Water in a garden has a special meaning in the Islamic culture, where water is very rare. It is a symbol for the paradise where waters flow without limitation.

Another features which are typical for an Islamic garden can also be found in the image: the geometric forms, the green grass and the lush green trees, the arches of the building and the almost graphic decorations on the building. Islamic culture does not allow pictures of persons or animals; Instead of this you will find beautiful calligraphy and tile work.

So a very simple prompt rendered an image which shows a typical islamic garden. Blue Willow always produces four images, and among those four the one above was the best.

From a simple prompt to a more complex one - Forest

 I do quite a bit of research about how to optimize my prompts. One advice is: Begin simple and then add to the prompt. Here is an example:

The prompt I began with is:

A forest at night with a full moon shining through the trees, fantasy, watercolor in the style of Anne Stokes, --ar 3:2

The AI Art Generator Blue Willow gave me the following image:

Nice, but I did not like the naked trees, and I also could do without the water in the middle.

So I changed the prompt to:

A dark green forest at night with a full moon shining through the trees, watercolor in the style of Anne Stokes, intricate, hyperrealistic, high definition

As you can see, the water disappeared and we have some green trees standing on green grass.

Now I wanted to add something fantasy like. What about some lanterns showing the way?

So I changed the prompt to:

A dark green forest at night with a full moon shining through the trees, red lanterns, mossy rocks, digital painting, intricate, hyperrealistic, high definition, 4 K

Here is the result of this prompt:

As you can see, we now have mossy rocks and red lanterns standing on the dirt.

Of course we could still go on with making the prompt more complex: We could add a woman with a long robe for example. And here the order of the words in the prompt is important: Everything which stands at the beginning of the prompt is considered to be most important to the AI Art generator. So one could put it at the beginning of the prompt - or more in the middle of it. 

And when you're done with prompting it could be helpful to write down the prompts and the way the image was rendered. There's a lot of trial and error in prompt crafting, especially when you try several AI Art Generator! Have fun!

Portrait of an Aztek warrior

 I love history, and I love to watch documentaries on tv about old civilizations, the Aztecs. AI Art provides me with the possibility to add images to my written knowledge. 

I learnt that the Aztecs were proud, cruel and effective warriors. With submitting a prompt to an AI Art Generator I managed to create the portrait of such a warrior very quickly. In spite of the short time it took generating the image provides a first impression of how these warriors could have looked like.

I also learnt that the Aztecs not only created beautiful pyramids but also big cities which were highly structured and well organized. 

Now I wanted to know how these pyramids could have looked and submitted a very simple prompt:

illustration of an aztec pyramid in Tenochtlitan

You can see the result on the right. 

So AI can be a tool to broaden our visual knowledge - or that of our children. At the moment AI is loved by techies, by people who are familiar with the internet and digital tools. However, the tools are getting simpler, which makes it easier for people who are not web-savvy to use these tools. 

Ravens

You can find ravens almost everywhere on the planet. They are highly adaptable and feed on almost anything, insects, seeds and fruits, carrion and small rodents.

They are known to be crafty: They have been watched carrying nuts to a busy street, letting them fall down on the tarmac. When cars pass over the nuts the nuts are cracked. The ravens come down when they see it safe and pick up the open nuts.

Besides being symbols of wisdom, trickery and mischief, ravens have been symbols of death and the afterlife: Some people think that something will happen to a member of the familiy or a dear friend if you see ravens flying over you.

Animal Wood Cut Prints

For me creating images with AI Art Generators is having fun but also learning about Art history and the various mediums applied in art throughout the centuries. I know quite a bit about art history in relation to painting, but not much about woodcut or linocut art.

The first thing I did was to ask Chatpgt for some information on wood cut printing. And it told me the following:

"Woodcut printing is a type of relief printing that involves carving an image onto a block of wood, applying ink to the surface, and then pressing the block onto paper or other materials. It is one of the oldest forms of printing and has been used in various parts of the world for thousands of years. 

In Europe, woodcut printing emerged as a popular medium for book illustrations during the 15th century."

Knowing all that I wanted to find out whether and how AI Art Generators would be able to create images looking like wood cut prints. I tried two of them:

1. Dreamlike AI

This dog woodcut print was created with Dreamlike AI. I like the way the fur and the background is rendered.

2. Blue Willow AI

This raven woodcut print was generated using Blue Willow. There is quite a difference in the style between the two images: Whereas the image above is really looking "old", the image on the right in my opinion looks rather modern. I used the same prompt; however dogs might have been quite often subjects for wood prints in the time of Albrecht Dürer, whereas ravens were not so much depicted on wood cuts at Dürer's time.

My first selfmade coloring page - colored

 

This is a coloring page I created using Blue Willow. In contrast to the one I created using Nightcafe Studio the image was without any colored pencils or other coloring media. So I simply had to get it into the proper format, a DIN A4 page. I did this using Canva. Then I downloaded the image as pdf which is easier to print, especially because my printer does not print anymore. It does scans, but it does not print. So I had to put it on a usb stick and have it printed out.

I mounted the print on cardstock, grabbed my alcohol based markers and began coloring. After months of creating art digitally, it was nice to have markers in my hand and to choose the colors!

I am always looking for ways to combine doing art traditionally and digitally - and creating one's own coloring pages and then doing the coloring work is definitely one of them!

Create your own coloring page using Nightcafestudio

For some months now I have been creating art using various AI Art Generators. However I have not forgotten my traditional art media - mostly watermedia and alcoholbased markers. So I decided to find a way to combine both: I wanted to create my own coloring pages using AI Art and then color the result with markers.

The first thing I did was to go to Nightcafestudio and submitted the following prompt:

"skull coloring page, no background, vector, monochrome, intricate, hyperdetailed"

The result was quite nice regarding the skull itself. However I did not like the colored pencils on the upper left and lower right of the skull. 

So I went to lunapic.com and had the background removed. 

You can see the final image below the first one.

The next steps will be:

1. upload the image to Canva 
   
   
2. select a DIN-A4 template, put the image there and resize it if necessary
   
   
3. download it as a pdf

Pandabear - various styles of cuteness

 This pandabear eating a watermelon was created with Dreamlike AI. I chose a cartoonish style with a bit of 3 D. This image could serve as a nice design for kids' products on a print on demand shop like Zazzle. Of course the background should be removed which can be easily done with a webtool called upscale.media.

The style of the image on the right is quite different. It still has - in my opinion - a bit of cuteness, but the pandabear character looks rather determined and more like the hero of a comic novel. It is definitely more cartoon-ish than the image on the top left.

It was created with Bluewillow AI.

Ladybugs

In many countries and cultures ladybugs are considered as a symbol of good luck and fortune. Together with chimneysweepers and piglets they can be found among New Year's Eve decorations in Germany. 
 

In some cultures, ladybugs are associated with love: If a ladybug lands on the hand of an unmarried woman she will fall in love soon with a true loving man.

Ladybugs are also important for people who do gardening or farming: They feed on aphids and thus keep crops from being eaten up by them.

AI Art and the beauty of musical instruments

When I was still working at the Berlin University of the Arts I was responsible for the Calendar of its website. During the winter and the summer terms this calendar is full of events, and many of these events are concerts given by the students and their professors. Every event was announced together with a nice photo which showed either the musical instruments or the groups who played them. I must say that it was not easy at all to have good photos.

Now with AI Art in full progress it is possible to create images which show the beauty of a musical instrument. So I tried to create a violin in Blue Willow, with a very simple prompt:

show me a photo of a beautiful violin, highdefinition, hyperrealistic, 4 K

And here it is!

Now, let's try another musical instrument, which is typical for an orchestra: a French Horn.

Here's the prompt:

Show me a photo of a musician playing the French Horn, intricate, hyperrealistic, highly detailed

You can see the result of the prompt on the right side. Unfortunately, the guy is not playing a French Horn. It looks more like a trombone. I tried more prompts, but obviously AI's database of images does not contain many images of musical instruments. Yes, it worked with the violin, but the violin is one of the most played instruments in orchesters.

Now I must say that I used Blue Willow with these prompts. There is still the question whether other AI Art generators like Bing Image Creator, Dreamlike AI, Lexica or Nightcafe Studio are better with creating images of French Horns, trombones, and other brass instruments.