Bing AI Image creator

Today I discovered the Bing AI Creator, an image creator powered by DALL-E, and I have to say that I love it. It is free. It works with so called boosting points which make the creation of your images relatively fast. You can earn additional boosting points by connecting your account to Microsoft Awards. 

I tried out a few things which I know that Stablediffusion has difficulties with. Here's the first one:

An octopus playing with a glass ball. You always get four images with your prompt, and I liked this one the best. 

I already have a selection of Initial Letters created with DALL-E2 over at Nightcafestudio. This one has the same quality.

Candyland

 This is one of four images I got when I submitted "Candyland" as a prompt on Dreamlike AI. Actually I was pretty surprised that one word created so many "sweet" images. 

Maybe I should add "castle" to the prompt, as in "show me the castles of candyland"? At the moment I am a bit down with credits on Dreamlike, so it has to wait...

Old Elves

Most of the elves you see on paintings, be they digital or not, are female; only some elves are male. In addition to that they are young adults or children. So I asked myself: What about old elves? Do elves die young? That made me thinking about creating old male elves - probably members of a kind of Elder Elves Council.

Here's the first old elf, Christofur. He is a wise old man, still eager to learn new things and to help the younger ones with his advice. 

This one here is called Philosopur, and he is the one who knows all about elv history and elv philsophy. He writes elv chronicles and is a guardian of the big elv library.

Both images were created using Stable Diffusion on Nightcafestudio.

A special kind of landscape

 After having used DALL-E2 on Nightcafestudio Art Generator for a long time I decided to focus more on Stablediffusion there. I tried to go for more painterly styles with landscapes, so I used the following prompt:

landscape with apple trees in full blossom, by artist sandro botticelli

I generated four images, and the image on the left is one of them. I like the fact that it does not look like a photo, with its rather stylized trees.

Using the same prompt on nightcafestudio and dreamlike.ai

 I have been using Nightcafestudio for a long time, and Deep Dream Generator for about a month. A few days ago I discovered dreamlike.ai, and I must say that I am quite impressed with the results I am getting with this AI Art generator. When you sign up for it you get 100 credits free; after that you can reclaim 25 credits every day. You can decide how many creations you want to get with one prompt - I usually go for 2 images which cost me 3,4 credits. 

Just now I am experimenting using a prompt from Lexica.ai with Nightcafestudio and Dreamlike.ai. Here is one example:

The prompt is:

ultra detailed tree in a generative 3D style on a disk, Grotesk font, Graphic Design, uppercase letter, fungal, mycelium, intricate concept art, triadic color scheme, by alphonse mucha and Mandelbrot, octane render, 4k, behance, dribbble, alphabet challenge

1. Nightcafestudio

And this is the image I got from Dreamlike.ai. I noticed, that the image on lexica (where I got the prompt from) looks quite similar. 

By the way, Dreamlike.ai is an AI art generator which allows the commercial use of creations without any limitations. Of course you have to enhance your creations in order to make it possible to use them for print on demand. 

Big Bad Wolf

 

I have been playing around with Stablediffusion on Nightcafestudio lately. In the meantime I've got so many credits that I can really experiment - adding a word here in the prompt, or taking away a special expression. 

I like the glowing eyes of the wolf on the left and the fur around them. He or she looks very determined, but not necessarily evil. 

Now, the wolf on the right looks very different. It looks like a creature of the night - evil. The color combination - black and a neon blue - add to this impression. 

Both wolf portraits have the same prompt - but the results are very different!

Brown bears

I love watching documentaries about wildlife, and I love watching documentaries about bears - big brown bears, grizzly bears, polar bears. I also did a few drawings of bears, mostly with markers. 

Since the end of the year 2022 I have been dabbling with creating AI artworks, using various AI Art generators - mostly nightcafestudio and deep dream generator. At the moment I am not so much interested in generating photorealistic images - I am trying to generate illustrations. 

The prompt for this bear is:

A brown bear in the forest, by Anne Stokes, watercolor and ink, comics color style, pen and ink

I very much like this illustration - it is not a photorealistic one but it renders this big animal very well.

I can see such an illustration in a children's book about bears, with information about how bears live.

However, I decided to generate another bear in a completely different style, a comic book style.

Here is the prompt for the image on the left:

Style: flat comic book style, clean ink lines, limited bold color palette with dramatic highlights, by Frank Miller, Akira Toriyama, Brian Bolland,

This could be an illustration out of comic novel, maybe telling a story about humans and a bear...

Welcome to fantasyland!

 

Here is Cyrus, the mighty red dragon. Well, he might not be so mighty, but he is red, and he really looks a wee bit frightening. He looks like defending something - the forest, or a dragon's nest, or whatever. 

Dragons usually have a bad reputation - they eat young inncocent maiden for breakfast, they are violent, they are evil. But: maybe they are just creatures of the wilderness and trying to defend nature against men who come into the forest and cut the wood down?

Dwarfs usually have a good reputation. They are workoholics, getting minerals out of the earth all day long, working with big hammers on metals. They usually come in groups (like the seven dwarfs). However, there might be a dwarf who is a bit special.

This one here seems to be quite young, and from the look of his clothes is not so much working in a mine, but seems to be living in the woods. And have a look at the look on his face! He looks very determined. It could be that he has discovered the red dragon and is very much determined not to give way. On the other hand he's got no weapons. Maybe he and the dragon find a way to get along with each other. Maybe he tells the dragon that he wants to defend nature as well, and so they are both on the same side.

Using the Clone tool with AI generated Images

 AI generators in general are not very kind to human limbs - and sometimes this also applies to animals. You can either spend all your energy points / credits or whatsoever, desperately trying to generate a human form with only two arms and not three - or you can try to correct the image with using the clone tool in Gimp or Photoshop. I have been using Gimp now for years, and in most cases it works as well as with Photoshop.

I generated this image with Deep Dream Generator. Unfortunately, the girl elf to the right has got two additional boots, and the girl elf to the left has got three hands. However, all in all I like the image.

So I downloaded it and openend it with Gimp. I know a friend of mine who can do anything with Photoshop, and I know that many images you see today in magazines are photoshopped. I have - until now - not touched the clone tool in Gimp. So I am an absolute newbie with it. But why not learning new things?

Here is the result: As I mentioned above, I am an absolute clone tool newbie - so the image is far from perfect. But as I intend to practise this tool more and more I will see progress!

Combining Nightcafestudio and Deep Dream Generator

Some of the results of my creation activity on Nightcafestudio are not fit to publish. However, I only throw images away when they are really, really bad. I keep some of the creations which I do not publish. 

Now, that I have reached the "Dreamer" status on Deep Dream Generator and therefore have more energy points to work with I asked myself: What would happen if I use those unpublished Nightcafe creations as base images with Deep Dream Generator?

As you can see, this image is not something you would publish. The ear of the elf on the right is far too big, and the hands are not good. (AI in fact has great difficulties with rendering hands - often they have too many fingers, or there is a third arm)

I uploaded the image above to Deep Dream Generator and used it as a base image, with the same prompt as on Nightcafestudio but adding  "watercolor and ink" to the prompt.

This certainly won't be the last experiment with Nightcafestudio and Deep Dream Generator!
 

Monster in a teacup - three variations

 Nightcafestudio has daily challenges. I participate only in the challenge that talks to me - in this case it is "cute animals in a teacup". I didn't want to add the 1002nd to the all the puppies, little cats, mice or whatever in a teacup, so I created a cute little monster.

Creation done in Nightcafestudio

It is cute, but I wanted to make more of it. Now, you can evolve images on Nightcafestudio with Stable Diffusion, but up to now the results that I got were not very satisfying for me. 

Deep dream Generator, the second AI Art Generator I have been using so far also works with so called "Base images". So I had a go with the image I created in Nightcafe, using the following prompt:

 "A cute monster looking out of a mug; colorful, in the style of Jean Baptiste Monge"

And here is the result:

Deep Dream Generator, 1st version

As you can see, the style of the image is very soft and fluffy, and the teacup is nicely decorated.

However, I wanted to have a third variation. This time the prompt was:

A cute monster with two paws looking out of a colourful teacup. Art deco, art nouveau, intricate, highdefinition, hyperrealistic

Deep Dream Generator, 2nd version

 You can see now the influence of the words "art deco, art nouveau" in the prompt. Even the teeth of the monster are decorated, and most of all the teacup.

Creating images with Lexica ai

 I havve been using three ai art generators so far:

1. Nightcafestudio
2. Deep Dream Generator
3. Lexica Ai

All of them can be used for free. Whereas Nightcafestudio and Deep Dream Generator are based on a credit /energy points system, Lexica offers you 100 images per month for free. This may seem to be quite a lot, but you have to keep in mind that every time you generate an image, you get four images. 

Here is an example:

You can easily see that you really get four different images with the following prompt:

jungle with moss covering the dirt, colorful flowers, trees with thick roots, in the style of Ivan Bilibin, Ernst Haeckel, Daniel Merriam, watercolor and ink

You can choose the dimensions, either square, portrait or landscape. I downloaded one of the four images and it had 2560 px width and height. After postprocessing and downscaling it looks like this:

As you can see, the style is a bit dreamy and fantasy like - but that's okay with me.

So I intend to continue to have fun with Lexica - having to watch myself not to create too many images!

Taking a photo as a starter image in Stable Diffusion

 

This is the starter image I used. The photo itself is not very good qualitywise (it once was a dia slide but then I had it digitalized, and as it shows a family trip to the Swabian Alb when I was a kid, I wanted to keep it), but it is nicely framed with the two big fir trees.

Now with Stable diffusion on Nightcafe, you can choose a starter image, and I had some kind of fantasy landscape in mind.

And this is one of the results. I had added "mountains in the background" in the prompt.

This is the second image, generated with the the same prompt. Stable diffusion added some water here without being told so, and instead of a tree on the left there is something like a rock, or parts of a mountain.

Both aiartworks have the same dreamy like, hazy feeling. 

The penguin with a scarf

This the story of penguin Jonathan.

He lived together with his fellow penguins on the antarctic coast. His mates were jumping in and out of the cold water, and when the mighty storms came, they huddled together. Okay, it was not that pleasant, but that's the life of a penguin, isn't it.

Jonathan always had been feeling cold. He was told by his parents to get himself together and play with his mates. He did, feeling miserable. 

Then one day, some humans approached the coast with boats and went ashore. They spent the whole day doing rather mysterious things, and then went off again over the sea. Slowly Jonathan and his mates went to the place where the humans  had been active. Suddenly, Jonathan saw something red lying on the snow.

He went near it and looked at it from a safe distance. Then he touched it very carefully with his beak. The red thing did not move. He touched it with one of his wings - and oh, it felt so soft and warm! He tried to wrap it around his neck, which was quite difficult and oh, he didn't feel cold anymore!

When he came back to his mates, they looked at him, shook their heads, and murmured something like "how unpenguinlike one can get!"Jonathan did not care. He now enjoyed playing penguin games like he had never done before!