I-DragGAN Interactive Editing
I-DragGAN ikuvumela ukuthi uhlele isithombe ngokuhudula amaphoyinti ngokoqobo: bamba indawo bese usihudulela lapho uqondiswe khona, futhi isithombe sonakala ngokwangempela, sishintsha ukuma, ukuma, noma indlela yokukhuluma.
Uhlolojikelele
It matters because it makes precise, intuitive image manipulation possible without sliders, masks, or text prompts.
I-Deep Dive
U-DragGAN, ovela ku-Pan, Tewari, Leimkuhler kanye nozakwabo bakwa-Max Planck kanye nozakwethu (SIGGRAPH 2023), wethule ukuhlelwa okusekelwe ephuzwini okusebenzisanayo kwezithombe ezikhiqizwe i-GAN. Umsebenzisi ubeka iphuzu elilodwa noma ngaphezulu 'lokubamba' esithombeni kanye nephoyinti 'eliqondiwe' elihambisanayo lapho kufanele bathuthele khona. I-DragGAN bese igudluza ngokuphindaphindiwe ikhodi efihlekile ukuze okuqukethwe ngaphansi kwesibambo ngasinye kushelele kuqondiswe lapho sonke isithombe sihlala sihlangene. Ungakwazi ukunweba imilenze yesilwane, wenze umuntu amamatheke, uzungezise imoto, noma ushintshe amakhonsathi ezwe, konke ngokuhudula. Okubaluleke kakhulu, ukuhlela kuhlonipha inqwaba yesithombe esifundiwe, ngakho imiphumela ihlala ingokoqobo kunokugcoba amaphikseli. Imaski ongayikhetha ikhawulela ukuthi yiziphi izifunda ezivunyelwe ukuhamba, inikeze ukulawula okuhle kwendawo.
I-Technical Insight
I-DragGAN isebenza esikhaleni esicashile se-GAN esiqeqeshelwe ngaphambili. Isebenzisa izinyathelo ezimbili ezishintshanayo: ukugadwa kokunyakaza, okushintsha ikhodi efihlekile ukuze izici eduze kwesibambo ngasinye ziye endaweni eqondiwe, nokulandelela iphoyinti, okuthutha isibambo ukuze silandele isici ebesigxiliswe kuso kusetshenziswa ukusesha komakhelwane oseduze kumamephu wesici. Ukuphinda lezi zinyathelo kuhambisa isithombe ngakunye kwe-GAN, kukhiqiza ama-deformation abushelelezi, namaqiniso.
I-Strategic Impact
Isivinini nesikali
I-Visual AI ingakwazi ukuhlola, ukutholwa, nokumaka imisebenzi esikalini.
Yakha ukukhetha
Amathimba aqanjiwe angakwazi ukulinganisa imiqondo ngokushesha ngezibuyekezo ezimbalwa ezenziwa mathupha.
Ithimba kanye nokusebenza komsebenzi
Imisebenzi ingasebenzisa amasiginali wesithombe nawevidiyo obekunzima ukuwenza ngaphambilini.
Ikusasa le-DragGAN Interactive Editing
I-DragGAN ivuse umsebenzi wokulandelela osheshayo iletha ukulawula okususelwe ekuhuduleni kumamodeli okusabalalisa (afana ne-DragDiffusion ne-FreeDrag), aphatha izithombe zangempela nokuqukethwe okungenangqondo kakhulu kunama-GAN kuphela. Lindela ukuhlela ukuhudula ukuze kube ithuluzi elivamile kusofthiwe yokudala, ehlanganiswe nezilawuli zombhalo neziyingi, futhi inwetshelwe kuvidiyo ne-3D ukuze abasebenzisi bakwazi ukumisa izinto kuwo wonke amafreyimu noma bamise kabusha amameshi ngokuhlanganyela, konke kuyilapho belondoloza i-photorealism.
Ukuqaliswa Komhlaba Wangempela
Ukulungisa indlela yokuma kwesithombe, indlela umuntu abuka ngayo, noma isitayela sezinwele ngokuhudula amaphoyinti obuso
Ukushintsha ukuma kwesilwane noma imoto nokuma kwayo, njengokuzungezisa imoto noma ukubeka kabusha ikhanda lebhubesi.
Ukulolonga kabusha izithombe zomkhiqizo (ukunwetshwa, ukunwetshwa, noma ukumisa kabusha izinto) kuma-mockups wedizayini
Ukulungiswa kahle kwezwe noma izithombe zemfashini ngokuhudula amakhonto, njengokushintsha izimo zentaba noma ukulingana kwengubo
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.
Hlola ngedatha efana nezimo zangempela zokukhiqiza.
Engeza isibuyekezo somuntu ukuze uthole ukuzethemba okuphansi noma izibikezelo zomthelela omkhulu.
Landelela ukukhukhuleka kwemodeli bese uqinisekisa kabusha ngemva kwezinguquko zekhamera noma zesethi yedatha.
Qhubeka Uhlole
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the DragGAN Interactive Editing quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Umhlahlandlela olandelayo
Tune-A-Video Ukuhlela Okukodwa
Imibuzo evame ukubuzwa
What is DragGAN Interactive Editing?
I-DragGAN ikuvumela ukuthi uhlele isithombe ngokuhudula amaphoyinti ngokoqobo: bamba indawo bese usihudulela lapho uqondiswe khona, futhi isithombe sonakala ngokwangempela, sishintsha ukuma, ukuma, noma indlela yokukhuluma. Ibalulekile ngoba yenza ukukhohlisa kwesithombe okunembayo, okunembile kwenzeke ngaphandle kwezilayidi, imaski, noma ukwaziswa kombhalo.
Umsebenzisi usihlela kanjani isithombe ku-DragGAN?
I-DragGAN ihlela umsebenzi ngokubeka amaphoyinti okubamba futhi uwahudule ukuze uqondise izindawo, ukhubaza isithombe ngokufanele.
Yiziphi izinyathelo ezimbili ezibalulekile ezishintshanayo ze-DragGAN?
I-DragGAN ishintshanisa ukugadwa kokunyakaza (izici ezihambayo ziye kokukhonjiwe) nokulandelela iphoyinti (ithola kabusha izibambo), iphindaphinda kuze kube yilapho kwenziwa.
Kungani ukuhlela kwe-DragGAN kubukeka kungokoqobo kunokugcotshwa?
Ngoba ukukhohlisa kwenzeka endaweni ecashile ye-GAN eqeqeshwe kusengaphambili, okuphumayo kuhlala kunqwaba yezithombe ezingokoqobo.
Ilawula ini imaski yokuzikhethela ku-DragGAN?
Imaski ivumela abasebenzisi ukuthi bakhawulele ukuhlela ezifundeni ezikhethiwe ukuze sonke isithombe sihlale singashintshi.
I-DragGAN yethulwe ngokuphawulekayo kuyiphi indawo ngo-2023?
I-DragGAN ishicilelwe ku-SIGGRAPH 2023, ingqungquthela ehamba phambili yezithombe zekhompyutha.