Uhlolojikelele
Amamodeli Okungaguquguquki Okufihlekile (ama-LCM) ayindlela evumela amajeneretha ezithombe ezisabalalisa ukuthi akhiqize izithombe zekhwalithi ephezulu ngesinyathelo esisodwa kuya kwezine esikhundleni senqwaba evamile. Benza ukukhiqizwa kwesithombe esiseduze nesikhathi sangempela, okusebenzisanayo kusebenze ngisho naku-hardware enesizotha.
Amamodeli Okungaguquguquki Okufihlekile awokugeleza komsebenzi okubonwa ngekhompyutha ahumusha noma akhiqize imidiya ebonakalayo ukuze ihlaziywe, isebenze, futhi isungule.
I-Deep Dive
Amamodeli ajwayelekile okusabalalisa acashile afana ne-Stable Diffusion aqala emsindweni futhi abuye abe umsindo ngokuphindaphindiwe, ngokuvamile adinga ukuhlolwa kwenethiwekhi okungama-20 kuya kwangu-50 ukuze enze isithombe esisodwa, esihamba kancane. Ama-LCM, athulwe uLuo kanye nozakwabo ngo-2023, asebenzisa i-distillation engashintshi endaweni ecashile yemodeli yokusabalalisa eqeqeshwe kusengaphambili. Umbono oyinhloko: qeqesha inethiwekhi yomfundi ukuthi igxumele ngokuqondile kumphumela ohlanzekile ukusuka kunoma iyiphi indawo eduze kwe-denoising trajectory, ngakho impendulo efanayo ifinyelelwa esinyathelweni esisodwa esikhulu esithathe eziningi ezincane ngaphambilini. Umphumela uba izithombe ezibukhali ezinyathelweni ezi-1 kuye kwezi-4. Indlela engumngane, i-LCM-LoRA, ipakisha lokhu kusheshisa njenge-adaptha encane ye-plug-in engadedelwa kumamodeli akhona ashunwe kahle we-Stable Diffusion ngaphandle kokuqeqesha kabusha inethiwekhi yonke.
I-Technical Insight
Amamodeli afanayo aphoqelela indawo 'yokuzinza': noma yimaphi amaphoyinti amabili endleleni efanayo yokukhipha umsindo (umzila wamathuba okugeleza kwe-ODE) kufanele imephu iye esithombeni esifanayo sokugcina esihlanzekile. Umfundi ukhishiwe kumodeli yokusabalalisa uthisha ukuze anelise lokhu, efunda ukubikezela isiphetho se-trajectory ngokuqondile. Ukusebenza endaweni efihlekile ecindezelwe esikhundleni samaphikseli kwenza ukucwiliswa kwamanzi kushibhe. Ngoba ukuhlola okukodwa kungagxumela ku-trajectory, isampula esindayo ephindaphindayo igoqeka ibe yizinyathelo ezimbalwa.
Ukuphatha Amamodeli Okungaguquguquki Okufihlekile
Amamodeli Okungaguquguquki Okufihlekile (ama-LCM) ayindlela evumela amajeneretha ezithombe ezisabalalisa ukuthi akhiqize izithombe zekhwalithi ephezulu ngesinyathelo esisodwa kuya kwezine esikhundleni senqwaba evamile. Benza ukukhiqizwa kwesithombe esiseduze nesikhathi sangempela, okusebenzisanayo kusebenze ngisho naku-hardware enesizotha. Amamodeli Okungaguquguquki Okufihlekile awokugeleza komsebenzi okubonwa ngekhompyutha ahumusha noma akhiqize imidiya ebonakalayo ukuze ihlaziywe, isebenze, futhi isungule. Ukuze wakhe ukuqonda okujulile, phatha Amamodeli Okungaguquguquki Okufihlekile njengemodeli yokusebenza, hhayi isici esisodwa: chaza imiphumela efiselekayo, ucacise ukucabanga, futhi uhlukanise lokho isistimu engakwenza ngokwethembeka kulokho okusadinga ukwahlulela kochwepheshe.
Empeleni, amaqembu aqinile asebenzisa amamodeli e-Latent Consistency ebhalansi ngokunemba namaqiniso okusebenza njengekhwalithi yedatha, ukuhluka kokukhanya, nokuvumelana kwamalebula. Babhala imibandela yempumelelo ecacile, ukuhlola okuqhathaniswa nedatha engokoqobo nokugeleza komsebenzi, futhi baphindaphinde ngokusekelwe kumaphethini okuhluleka aqashiwe esikhundleni sokuwina kwebhentshimakhi yesikhathi esisodwa. Yilapho ukuqonda kwethiyori kuguquka kube amandla ahlala njalo kuwo wonke umkhiqizo, inqubomgomo, kanye nokusebenza.
I-Visual AI ingakwazi ukuhlola, ukutholwa, nokumaka imisebenzi esikalini. Ngesikhathi esifanayo, amalungelo ezithombe kanye nemvume kungaba ubungozi bomthetho uma ukutholakala kungacacile. Indlela eqine kakhulu iwukuhlanganisa isivinini sokuhlola nesiyalo sokuphatha: qhuba abashayeli bezindiza, bamba ubufakazi, ushicilele amalogi ezinqumo, futhi ubuyekeze izivikelo ngokuqhubekayo njengoba imodeli yokuziphatha, okulindelwe ngabasebenzisi, kanye nezimfuneko zokulawula zishintsha.
I-Strategic Impact
I-Visual AI ingakwazi ukuhlola, ukutholwa, nokumaka imisebenzi esikalini.
I-Visual AI ingakwazi ukuhlola, ukutholwa, nokumaka imisebenzi esikalini. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.
Amathimba aqanjiwe angakwazi ukulinganisa imiqondo ngokushesha ngezibuyekezo ezimbalwa ezenziwa mathupha.
Amathimba aqanjiwe angakwazi ukulinganisa imiqondo ngokushesha ngezibuyekezo ezimbalwa ezenziwa mathupha. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.
Imisebenzi ingasebenzisa amasiginali wesithombe nawevidiyo obekunzima ukuwenza ngaphambilini.
Imisebenzi ingasebenzisa amasiginali wesithombe nawevidiyo obekunzima ukuwenza ngaphambilini. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.
Ukuqaliswa Komhlaba Wangempela
Amathuluzi ekhanvasi esikhathi sangempela abuyekeza isithombe esikhiqiziwe njengoba uthayipha noma udweba, ngokunensa okungazero
Isebenzisa ukukhiqizwa kwesithombe sokusabalalisa okuzinzile kukhompuyutha ephathekayo noma i-GPU yefoni engxenyeni yesekhondi
Idedela i-adaptha ye-LCM-LoRA kumodeli ekhona eshunwe kahle ukuze isheshiswe ngokushesha ngaphandle kokuqeqeshwa kabusha
Ukukhiqiza amaqoqo amakhulu ezithombe ngemali ephansi ukuze kuhlolwe idizayini ngokusika izinyathelo ukusuka ku-~30 kwehle kuye kokuthi ~4
Amaphethini Okusebenzisa
Amamodeli Okungaguquguquki Okufihlekile ekusebenzeni
Amathuluzi ekhanvasi esikhathi sangempela abuyekeza isithombe esikhiqiziwe njengoba uthayipha noma udweba, ngokunensa okucishe kube uziro.
Amathuluzi ekhanvasi esikhathi sangempela abuyekeza isithombe esikhiqiziwe njengoba uthayipha noma udweba, cishe nama-zero lag Amaqembu ngokuvamile athola imiphumela engcono uma echaza izinga eliphezulu ngaphambili, egcina indlela yokukhuphuka yomuntu yamakesi asemaphethelweni, futhi alandelele kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.
Amamodeli Okungaguquguquki Okufihlekile ekusebenzeni
Isebenzisa ukukhiqizwa kwesithombe sokusabalalisa okuzinzile kukhompuyutha ephathekayo noma i-GPU yefoni engxenyeni yesekhondi.
Ukuqalisa ukukhiqizwa kwesithombe Se-Stable Diffusion kukhompuyutha ephathekayo noma i-GPU yocingo engxenyeni yesekhondi Amaqembu ngokuvamile athola imiphumela engcono uma echaza imingcele yekhwalithi ngaphambili, egcina indlela yokukhuphuka yomuntu yamakesi asemaphethelweni, futhi alandelele kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.
Amamodeli Okungaguquguquki Okufihlekile ekusebenzeni
Idedela i-adaptha ye-LCM-LoRA kumodeli eshunwe kahle ekhona ukuze isheshise ngokushesha ngaphandle kokuqeqeshwa kabusha.
Ukudedela i-adaptha ye-LCM-LoRA kumodeli evele icushwe kahle ukuze isheshiswe ngokushesha ngaphandle kokuqeqesha kabusha Amathimba ngokuvamile athola imiphumela engcono lapho echaza imingcele yekhwalithi ngaphambili, egcina indlela yokukhuphuka yomuntu yamakesi asemaphethelweni, futhi alandelele kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.
Amamodeli Okungaguquguquki Okufihlekile ekusebenzeni
Ukukhiqiza amaqoqo amakhulu ezithombe ngemali ephansi ukuze kuhlolwe idizayini ngokusika izinyathelo ukusuka ku-~30 kwehle kuye kokuthi ~4.
Ukukhiqiza amaqoqo amakhulu ezithombe ngemali eshibhile ukuze kuhlolwe idizayini ngokusika izinyathelo ukusuka ku-~30 kwehle kuye kokuthi ~ 4 Amaqembu ngokuvamile athola imiphumela engcono uma echaza izinga eliphezulu ngaphambili, egcina indlela yokukhuphuka yabantu yamakesi asemaphethelweni, futhi alandelele kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.
Izingozi & Guardrails
Amalungelo ezithombe kanye nemvume kungaba ubungozi bezomthetho uma ukuvela kungacacile.
Ukusebenza kwemodeli kungahluka kukho konke ukukhanya, izibalo zabantu, kanye nezindawo.
Okuhle okungelona iqiniso kungase kungabonakali ngaphandle uma izinga lokuzethemba liqashelwa.
Ukuqalisa Umhlahlandlela
Chaza indlela yokwamukela yokunemba, ukukhumbula, nezindleko zamaphutha.
Chaza indlela yokwamukela yokunemba, ukukhumbula, nezindleko zamaphutha. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.
Hlola ngedatha efana nezimo zangempela zokukhiqiza.
Hlola ngedatha efana nezimo zangempela zokukhiqiza. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.
Engeza isibuyekezo somuntu ukuze uthole ukuzethemba okuphansi noma izibikezelo zomthelela omkhulu.
Engeza isibuyekezo somuntu ukuze uthole ukuzethemba okuphansi noma izibikezelo zomthelela omkhulu. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.
Landelela ukukhukhuleka kwemodeli bese uqinisekisa kabusha ngemva kwezinguquko zekhamera noma zesethi yedatha.
Landelela ukukhukhuleka kwemodeli bese uqinisekisa kabusha ngemva kwezinguquko zekhamera noma zesethi yedatha. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.