HejuruUbuyobozi bukurikira
Icyitegererezo cyo Gutandukanya
AI igaragara
Amabwiriza ya AI
GLIDE yari kare OpenAI icyitegererezo cyo gukwirakwiza inyandiko-yerekana ishusho yerekana ibisobanuro wongeyeho 'ubuyobozi butarangwamo ubuyobozi' bushobora gutsinda sisitemu ya GAN mbere.
Umva iki gitaboAI-byavuzwe · ~2 min
It was a key stepping stone on the path to DALL-E 2.
Released by OpenAI in late 2021, GLIDE (Guided Language to Image Diffusion for Generation and Editing) demonstrated that diffusion models guided by text could produce photorealistic, prompt-faithful images. Its biggest contribution was comparing two ways to steer generation: CLIP guidance versus classifier-free guidance. The team found classifier-free guidance produced more realistic and better-aligned images, a result that shaped nearly every text-to-image model since. GLIDE also supported text-driven inpainting, letting users edit part of an image with a new prompt. It used a 3.5-billion-parameter diffusion model plus an upsampler. OpenAI released a smaller, filtered version publicly while withholding the full model over misuse concerns, and the DALL-E 2 research paper describes an unCLIP decoder that modifies the GLIDE architecture to accept CLIP-image conditioning.
Gukora ishusho kuva mu nteruro nkibintu byasobanuwe, byerekana hakiri kare-kwizerwa
Gushushanya inyandiko-gushushanya: guhisha igice cyifoto no kuyuzuza ikintu gishya cyasobanuwe mumagambo
Guhindura ishusho iriho wongeyeho cyangwa usimbuze ibintu ukoresheje gukurikirana ikibazo
Gukora nkibanze shingiro ryubushakashatsi bwerekanaga ibyiciro-bidafite ubuyobozi bikubita ubuyobozi bwa CLIP kugirango uhuze
GLIDE ubwayo ni amateka cyane, yasimbuwe na DALL-E 2, Imagen, na Diffusion Stable, ariko ibitekerezo byayo bikomeza hose. Ubuyobozi budafite ubudahangarwa bukomeza kuba ipfunwe ryo gucuruza ubudahemuka no gutandukana, kandi inyandiko-yashushanyijeho ubu ni ibisanzwe. Sisitemu yigihe kizaza ikomeza kunonosora ingengabihe yubuyobozi, kugabanya ibihangano bikomeye bitera ubuyobozi, no kwagura amahame amwe kuri videwo na 3D ikwirakwizwa, bityo imbaraga za GLIDE zirenze icyitegererezo.
Ibaruwa y'amakuru y'ubuntu
Inkuru eshatu zizewe za AI buri gitondo cy'iminsi y'akazi, zanditse mu rurimi rworoshye. Ubuntu iteka ryose, nta kwamamaza.
Imeri imwe buri munsi y'akazi. Kwiyimura mu buryo bumwe. Ntitwigera tugurisha cyangwa ngo dusangire aderesi yawe.
Isuzume
Ibisubizo ako kanya kuri buri kibazo, n'icyemezo gishobora gusangirwa gifite ID yagenzurwa umaze gutsinda isomo.
Fasha uburezi bwa AI bw'ubuntu. AI Understanding ni umuryango utegamiye kuri leta wa 501(c)(3) — nta kwamamaza, nta kwishyura, iteka ryose. Tanga impano
GLIDE yari kare OpenAI icyitegererezo cyo gukwirakwiza inyandiko-yerekana ishusho yerekana ibisobanuro wongeyeho 'ubuyobozi butarangwamo ubuyobozi' bushobora gutsinda sisitemu ya GAN mbere. Byari urufunguzo rwibanze rwinzira igana DALL-E 2.
GLIDE yerekanye ibyiciro-bidafite ubuyobozi byatanze ibisubizo bifatika kandi bihujwe neza kuruta ubuyobozi bwa CLIP.
The full research model used a 3.5-billion-parameter base diffusion model with a separate upsampling stage.
Muguhugura kubintu byombi kandi byubusa, icyitegererezo cyiga ibisekuruza bisabwa kandi bitagabanijwe, hanyuma bigahinduka hagati yabyo kuri sampling.
The paper describes adding CLIP-image conditioning to a modified GLIDE architecture for its unCLIP decoder; this is a concrete architectural relationship.
OpenAI yahagaritse icyitegererezo cyuzuye kubera impungenge mbi hanyuma asohora akayunguruzo, ntoya aho.
Komeza wige
Abandi bayobozi batoranijwe kuriyi ngingo
HejuruUbuyobozi bukurikira
Icyitegererezo cyo Gutandukanya
AI igaragara