1/15
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
|---|
No study sessions yet.
Graphic design
Creating images, postors, or layouts using art and computers
Illustrate
To show or explain something using pictures or drawings.
Suitable
Good or right for a specific purpose.
AI
A computer system that can think and learn like a human.
Neural network
A computer model designed to work like a human brain
Algorithm
A set of steps a computer follows to solve a problem.
Generative
Relating to creating or producing something new
Training set
A collection of data used to train ai.
Diffusion
A process of adding noise to a image and training ai to reverse it.
GAN (Generative Adversarial Network)
A way of using two ai models to generate something
What are some examples of things that AI can generate? |
AI can generate building video consequences or anywhere humans come up with a prompt and AI generates it. |
How did Generative Adversarial Networks make realistic-seeming images? |
Generative Adversarial Networks make realistic images by using a generator and an adversary. The generator generates images of what is imputed into the network and the adversary takes these images and checklist to see if they are real or generated and they continue to do this until they cannot tell if it is generated of real
What is happening in each version of the image? |
In each version of the image the computer makes the image more noisy until you only see noise
How does the diffusion process train computers to build new images with specific details?
The diffusion process trains computers by taking an already clear picture with many details and making it gradually more noisy until it is all noise. This helps the computer learn how to reverse this process and make the image clear and less noisy.
Personal opinion: who do you think should "own" the images that AI creates?
The programmer should own the images because they put in the program to generate them.
What can you imagine creating with a diffusion model?
I can imagine creating a meme from a diffusion model. I would write a prompt to generate an image.