//=time() ?>
Voilà 1000 dessins de robots créés par une IA, à partir d'une phrase différente à chaque fois. On est au bord d'un gros changement dans la façon de créer, tout ça va impacter les métiers de l'image (dont le JV) et leurs processus de création
▶️https://t.co/FwqvqJ9Iro
#dalle2
Pirate Ship Series
Prompt: Epic Pirate Ship flying over a Cyberpunk City Landscape
Made with LatentDiffusion model (not #dalle2)
A Cloudy Day in a Forest Valley
Made with #latentdiffusion not #DALLE2 series.
@WendiYan5 Dalle2 approached the space from a very technical viewpoint, so the images that are shared have been very performance focused (e.g. look it’s smart enough to represent X concept) but I think artists will wield it to make new aesthetics. Here’s some weird aesthetic I just made
#dalle2 has great coverage of the entire prompt. Instead of smushing all the nouns together, it includes most of the details.
"Dolphins with opposable thumbs discover tools"
A comparison of generative #art by Dall-E 2 model (left) and Latent Diffusion LAION-400M (right) for the same input prompt.
“A rabbit detective sitting on a park bench and reading a newspaper in a victorian setting” #dalle2 #laion400m #TextToImage