So, I searched for 'Lucky in the Sky with Diamonds ' with VQGAN + CLIP. Will image generation AI change the way art works? -GIGAZINE You can understand the combined attempt by reading the following articles. VQGAN is an image generation model that develops 'VQ-VAE ', which is a VAE using vector quantization (VQ), into a hostile generation network (GAN), and a neural network 'CLIP' that connects text and images to this. The related elements were not drawn in the image, probably because Meta changed the company name and it was a short time.Īccording to the tips on the site, if it is a long phrase that combines multiple words instead of a short keyword, it is better to change it to 'VQGAN + CLIP' instead of 'Hypnogram'. The 'A' on the left may be somewhat closer to Amazon's A. The impression that the Google and Apple logos are output fairly firmly. This time, I entered 'Google Amazon Meta Apple' and the name of a giant tech company to create an illustration. And it is a mystery that a man's face still exists in the center of the image. The color is more gorgeous than the previous illustration, and the character string in the center of the screen is closer to the GIGAZINE logo. Below is another illustration created by AI from 'GIGAZINE'. Looking at the illustrations, somehow two 'GIGAZINE' logo-like things are drawn, and I'm curious that the illustration of only the lower half of the man's face is attached for some reason.Īn illustration is created each time you click 'Generate', so a different illustration will be created each time. When the illustration is completed, the illustration will be displayed below the input field, and you can save the illustration in 480 x 480 pixel JPEG format by clicking 'Download'. This time, I asked AI to create an image with 'GIGAZINE'. However, it takes a long time to process, so you need to wait about 2 to 3 minutes. To create an image, just enter a keyword and click 'Generate'. I tried various keywords to see what kind of picture would actually be drawn. 'Hypnogram' is a site that automatically draws illustrations imaged by AI from the entered keywords, and can be used from the browser. 20:00:00 'Hypnogram' that draws an illustration imaged from the keyword entered by AIĪdvances in technology have also made it possible for algorithms learned in deep learning to draw pictures.
0 kommentar(er)
