Yellow Roses

Yellow Roses - lineart

I love roses, especially yellow roses. I got some very nice yellow roses with DALL-E2 on Nightcafe. However, I wanted to achieve a more graphic look. So I added the word "lineart" to the prompt which now read:

A bouquet of beautiful yellow roses in a white vase, lineart, highdefinition, photorealistic, intricate

This prompt not only added lines to the yellow petals (which I had intended) but also some lines to the vase (which was not intended, but okay). 

 

Creating prints off your AI Images with Canva

 For some time now I paint and draw - with brush and paint and in a digital way, mostly with Inkscape. But only since the last two years or so I had my paintings printed out, framed and hanged on the wall or printed and used for greeting cards. I scan my pieces and have them printed in a nearby drugstore. 

Now, with my AI artworks it is a bit difficult, most of all when working with DALL-E2. DALL-E2 renders square images. Greeting card formats are rectangles, not squares. Blowing up the image takes parts of the image away which is not good. But there is a solution: Canva!

Here are some examples of what I did:

The images consist of three layers: 

• The first layer which later serves as a kind of frame is the background. In Canva, you can have backgrounds in the color you like, there are even backgrounds with textures. 
• The second layer is a white rectangle (rectangles can be found among "elements"). You can also have all kinds of color, but white resembles a mount. 
• The third layer is the image itself. 

Each layer can be positioned in the center and the middle.

When this is done, the image can be downloaded and be printed at the drugstore or wherever you go when you want to have your photos printed out.

Using Lexica in order to create images with DALL-E2

 Lexica is a website where you can do two things:

• Generate images with Stable Diffusion
• Search for images and get the prompts which created them. 

On Lexica, you have to pay in order to be able to create a certain number of images (Mind you, Nightcafe is free, but you have to do something in order to get credits for your creations - log in daily, vote in challenges, put your images on instagram and twitter). The  most important difference for me, however, is that you can upscale your images in Nightcafe,- paying with credits, and you cannot in Lexica. So I will not use Lexica for creating images.

However: Browsing through the images in Lexica gives you hundreds of prompts. In many cases the prompt and the generated image differ. I tried out some prompts in DALL-E2 on Nightcafe - and the result was more similar to the prompt than the image on Lexica itself. 

Here's an example:

I copied the prompt from Lexica and changed it a bit. It now reads:

Photo of a golden statue of a lion with huge wings

The original prompt included "a collar", which I did not want in the image.

Resources for DALL-E2 Part 1

 There are some really helpful resources for generating AI artwork with DALL-E2. One of them is  Dall-E wizard. You can find tutorials, guides, prompts and a gallery on this site. Prompts can be listed in an alphabetical order. Going through the list of prompts I found "kaleidoscope" - which inspired me to create the following image with the prompt:

kaleidoscope, blue and orange, digital art. trending in artstation

Of course you can change the prompt: choose other colours for example.

Playing with presets in DALL-E2 on Nightcafe

Presets are a set of modifiers which have an influence on the outcome of your image. On Nightcafe there a quite a few of those presets, They have names like "nightcafe", "epic", "dark fantasy" or "fantasy", just to name a few.

I'll give you an example:

Here's the image I started with:

Prompt:

medium close up marmor statue by artist Michelangelo of a Majestic wolf standing in front of dark mountains, photorealistic, intricate, hyperdetailed, 4 K, trending on Artstation

Nice, but a bit boring. So I decided to duplicate the image and add the preset "fantasy" to it. And here's the result:

And what does the prompt with the added preset look like?

medium close up marmor statue by artist Michelangelo of a Majestic wolf standing in front of dark mountains, ethereal fantasy hyperdetailed mist Thomas Kinkade

The red text is the text provided by the preset. It's the mist which provides a bit of mystery for the image...

Easter bunny - from pixar style to glass bunny

 Easter is still some weeks away, but nevertheless I felt like creating an easter bunny. So I opened DALL-E2 on nightcafe and inserted the following prompt:

photo of a cute laughing Easterbunny pixar style, highdefinition, 8 K, trending on artstation standing on green grass

and I got that image:

Well, isn't that a cheeky laugh? Probably the guy is laughing about silly hens which didn't look after their eggs...

So I thought about transforming that guy into something of an Easter decoration. I changed the prompt into:

blue glass sculpture of a cute laughing Easterbunny pixar style, highdefinition, 8 K, trending on artstation

and got this little bunny glass statue:

By the way, I tried the same prompt with Stable Diffusion 1.5 and I thoroughly failed. ...

Evolving an image in Nightcafe

 Actually there are two methods of evolving:

1. Evolving in DALL-E2: You copy your image with the existing prompt and add text or remove text.
2. You create an image with DALL-E2 and evolve it with Stable Diffusion. I do it when I notice that DALL-E2 doesn't handle a subject as well as Stable Diffusion would. In that case, you provide Stable Diffusion with a socalled starter image.

Here is an example:

I created these four images in DALL-E2:

As you can see, all of them have certain flaws: either the crowns do not look well on the heads, or the faces themselves have flaws. 

From looking at other people's images on Nightcafe I noticed that Stable Diffusion seems to render portraits better than DALL-E2 in many cases. So I decided to evolve one of the images. I clicked on the image I wanted to evolve and then on the evolve button. Now I had to change the creation method from DALL-E2 to Stable Diffusion. 

Here I changed the prompt a bit, adding "photo of a beautiful young woman" in order to get a decent looking face. I then chose 4 images and clicked on Create. And here's the result:

It looks quite different than the starter image, doesn't it?

Getting the result you want step by step with DALL-E2

 The first step you have to do is: Visualize the desired result. You have to have an image in your head - be it a dog running over a green meadow, a vase with tulips, whatsoever. I have been painting a lot of masks, and I am almost obsessed with masks. Masks however is such a general notion that it has to be specified. 

And that is the second step: Be as precise with the image in your head as possible: African masks? Masks from other continents? I wanted an image with Venetian masks - they are so pretty, they are the symbol of carnival in Italy. So I wrote the following prompt in DALL-E2:

Photo of a Venetian mask, medium closeup, photorealism, hiyperdetailed, intricate back lit 4K

And here's the Image:

Mask 1

Well, the image is not bad, but not what I had in mind when I was thinking of Venetian Masks. So I changed the prompt to:

Photo of a Venetian mask, photorealism, hiyperdetailed, intricate, 4K

As you can see when comparing the two prompts I added something and I deleted something.  This resulted in:

Mask 2

Let me add one important information here: I always let DALL-E2 create four images, and up to now I could use one or two out of these four. 

So Step #3 means: try to adapt the digital image to the image in your head by working on the prompts. Generating images with DALL-E2 is not something like hitting the "Create" button and there it is - the beautiful fantastic extraordinary image you had in your mind! It means spending time - and in regard to Nightcafe - credits. It is a process. 

The Magician

 Nightcafe Studio runs daily challenges, and the challenge of 10 January 2023 was "Magicians". 

I wanted to participate in this challenge, so I opened DALL-E2 and inserted the following prompt:

"Photo of a magician in a long blue robe, with a long white beard, standing in front of a fireplace, reading an old book"

Here's the result:

Well, the magician looks more like a nice elderly man in a cosy blue bathrobe, reading out of a fairy tale book.Not  necessarily that what I intended.

So I changed the prompt a bit. It now reads:

"photo of a magician in a long blue robe and a black magic hat , with a long white beard, standing in front of a fireplace, reading an old book, gloomy atmosphere, hyperdetailed, photorealistic, intricate"

And yes, this prompt created some changes!

The uses of AI generated artwork

 

AI generated art can be used for many things:

1. it can be used for creating calendars and cards; I have already created the January page of one of my calendars with two prints (Winter in the forest); in addition to that I used images for some of my Christmas cards and will use them for birthday cards in 2023.
2. it can be used as reference for traditional drawing and painting.
3. it can be used for print on demand. I am the owner of a Zazzle shop, and a few days ago I created my first mug design with it - a baby rhino.  

Baby Rhino
by ullahennig

Nightcafestudio: Working with DALL-E Part 2

 Okay, you are on the Nightcafe website and you want to create an image with DALL-E2. Click on "Create". This will open a page with all the possible creation methods: Stable Diffusion, Dall-E2, Coherent, Artistic and Style Transfer. Today we will generate an image with DALL-E2, so click on the image with the cool green chairs. 

Now you have to insert a text prompt. Try to describe the image you want to create as precise as possible.

I chose: "Pixar style 3D render of a penguin wearing a tophat, 4k, high resolution, trending in artstation"

Now you have to decide how many images you want to create. Each image uses up 1 credit, so you may want to save credits by going for 1 image. But: the one image you go for may be awful. So I recommend going for 4 images, and then selecting the one you like best.

I got those four images:

All four penguins are cute, but I decided to publish the one in the lower right corner.

DALL-E2 does not have additional parameters to influence the outcome of the creation process, like Stable Diffusion has. There is only the text prompt. This can make it more simple, but also more complicated.

Nightcafestudio: Working with Dall-E2

 I've been experimenting on Nightcafe Studio for quite some time now (two months, I think), and I have been using Stable Diffusion, one of Nightcafe's algorithms, for most of the time. However lately I began to experiment with Dall-E2. It works a bit different, and it has less ways to manipulate the outcome. But it can understand things like "close up", "extreme close up and silmilar photography words.

Here are two results:

Elephant # 1 with the prompt: Close up elephant at the lake 4K hyperrealism HDR

and here's elephant #2:

I used the prompt: Medium Close up elephant at the lake 4K hyperrealism HDR

I must say, I am quite pleased with the outcome!

Nightcafe: Two versions with the same prompt

The generation of images in Nightcafe studio is managed by various parameters:

1. The text prompt
2. The version of Stable Diffusion
3. The Sampler
4. The Seed number

By the way: If you ask me, what parameters 2, 3 and 4 are - I don't know it. I am not a Techie. In Nightcafe you can select vario

us versions, samplers and you can change the number of the seed.

Let me give you an example:

The image below was generated with

• the text prompts: "castles on the moon with spaceships" and "neo-impressionism expressionist style oil painting, smooth post-impressionist impasto acrylic painting, thick layers of colourful textured paint"
• Stable Diffusion 1.5
• The Sampler K_LMS
• and the Seed number: 2127287723

And here comes the 2nd version of the image:

The text prompts and the Sampler are the same. But:

1. It is Stable Diffusion 2.0
2. Seed number is 2127287720

And what a difference between those two images! I cannot say which of them I personally prefer. I really like both of them.

So: Images in Nightcafe or any other AI tool are not generated by just hitting the button "create". Often it is a very time-consuming and tedious process!