AI and it's Art
I am the first to admit that I am oblivious to a lot of what is happening on Vocal. There is a lot to keep up with. I did ask to be made to feel warm and fuzzy, but now I feel like I if I am not careful, I will internally and externally combust. Things have certainly heated up on the platform.
atOptions = {
'key' : '82e8db775a0372a86fa77979cd2db96c',
'format' : 'iframe',
'height' : 90,
'width' : 728,
'params' : {}
};
document.write('
I still have right hand in and left hand out with Discord. Then there are the Facebook groups, which are getting more and more complex as one tries to navigate them. But, I am doing my best and working at my own pace, which leaves me miles behind everyone else. Totally fine, I ain't in no race with nobody.
atOptions = {
'key' : '82e8db775a0372a86fa77979cd2db96c',
'format' : 'iframe',
'height' : 90,
'width' : 728,
'params' : {}
};
document.write('
I realize that whenever I ask for just plain creations, with no pretty words attached, I tend to get what I ask for. For example, if I do not specify ethnicity, I will most assuredly get white folks all the time. White women tend to come out looking quite lovely. No offense, I love white folks as much as every other color folks. But I have to be very specific with what I request, in order to get what I almost want. 'Cause it is a toss up about that......
If I ask for "A pic of black women surrounded by flowers". I get this:
atOptions = {
'key' : '39ce2024dfc9b19ad640069e13df2956',
'format' : 'iframe',
'height' : 250,
'width' : 300,
'params' : {}
};
document.write('
atOptions = {
'key' : '39ce2024dfc9b19ad640069e13df2956',
'format' : 'iframe',
'height' : 250,
'width' : 300,
'params' : {}
};
document.write('I also notice that AI tends to expose or highlight Black women's breasts way more often than it does other (I won't mention who or whom else's) breasts. Case in point:
Black witch
White witch
If I ask for a pic of an older woman however, AI is very respectful of this and dresses her in matronly clothing with a most genteel aura.
Makes me wonder where AI draws its images from. Is it free to roam the worldwide web for pictures, or is it specifically programmed to draw its characters from specific written lines of code or whatever one refers to for images.
So, I say to you, be careful what you ask AI for, for you shall surely have it. As I told Mike, or was it Paul? You have to have a romance with AI. Whisper sweet nothings in his, her, its ears, and you will receive the most lovely of creations.
I asked Open AI for a picture of a pirate ship. I got this:
I have no idea how the guys created lovely centaurs for the centaur challenge, but they were gorgeous male and female centaurs.
Maybe I should have just asked for the body of a horse with a beautiful female head attached...Lord knows what I would have got!
I asked for: A centaur with the body of a horse and the head of a beautiful woman (NightCafe)
Imagery (Wiki)
"Many mechanisms for creating AI art have been developed, including procedural "rule-based" generation of images using mathematical patterns, algorithms which simulate brush strokes and other painted effects, and deep learning algorithms, such as generative adversarial networks (GANs) and transformers".
One of the first significant AI art systems is AARON, developed by Harold Cohen beginning in the late 1960s at the University of California at San Diego. Cohen developed AARON with the goal of being able to code the act of drawing. In its primitive form, AARON created simple black and white drawings. Cohen would later finish the drawings by painting them. Throughout the years, he also began to develop a way for AARON to also paint. Cohen designed AARON to paint using special brushes and dyes that were chosen by the program itself without mediation from Cohen
Generative adversarial networks (GANs) were designed in 2014. This system uses a "generator" to create new images and a "discriminator" to decide which created images are considered successful. DeepDream, released by Google in 2015, uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia, thus creating deliberately over-processed images.
I still don't know where it draws its images from after this well-informed read from Wikipedia, apparently it is a very convoluted network of neural interfaces and whatnots.
But I am getting wise in the ways of Ai art, peeking at and copying from the masters of Ai art on the various platforms. Most of them don't mind, they invite us to share and learn.
AI has come long way. Can we keep up?
.............................................................................................
Several companies have released apps that transform photos into art-like images with the style of well-known sets of paintings. The website Artbreeder, launched in 2018, uses the models StyleGAN and BigGAN to allow users to generate and modify images such as faces, landscapes, and paintings.
Several programs use text-to-image models to generate a variety of images based on various text prompts. They include VQGAN+CLIP which was released in 2021. OpenAI's DALL-E which released a series of images in January 2021, Google Brain's Imagen and Parti which was announced in May 2022, Microsoft's NUWA-Infinity, and Stable Diffusion which was released in August 2022. Stability.ai has a Stable Diffusion web interface called DreamStudio. Stable Diffusion is source-available software, enabling further development such as plugins for Krita, Photoshop, Blender, and GIMP, as well as the Automatic1111 web-based open source user interface.
There are many other AI art generation programs including simple consumer-facing mobile apps and Jupyter notebooks that require powerful GPUs to run effectively
Comments
Post a Comment