Eksempler på mine værker:GANs (Generative Adversarial Networks) are systems where two neural networks are pitted against one another: a generator which synthesizes images or data, and a discriminator which scores how plausible the results are. The system feeds back on itself to incrementally improve its score. CLIP (Contrastive Language–Image Pre-training) is a companion third neural network which finds images based on natural language descriptions, which are what’s initially fed into the VQGAN.
A lot of coverage has been on the unsettling and dystopian applications of GANs — deepfake videos, nonexistent but believable faces, poorly trained datasets that inadvertently encode racism — but they also have benign uses: upscaling low-resolution imagery, stylizing photographs, and repairing damaged artworks (even speculating on entire lost sections in masterpieces).
"Bruce Springsteen eating a scorpion" resulterer i dette:
"Mario fighting Luigi" resulterer i dette:
Og min favorit; "A wedding between a dove and a rabbit":
Og så tænkte jeg, at I - svingninger.dk - burde være stedet, der kan komme på fede idéer til blæste AI-billeder. Skriv en ganske kort engelsk beskrivelse af det billede, du ønsker at frembringe, og så smider jeg det i maskinen.
