Xiga xigaHagaha xiga
Kacdoonnada kala sooci kara ee qotodheer
Farsamo
HAGAHA Farsamada
Dilated and Arous Convolutions
Isku-dhafka kala-goysyada (sidoo kale loo yaqaan convolutions atrous) waxay gelisaa farqiga u dhexeeya miisaanka shaandhada si kernelku u daboosho aag aad u weyn iyada oo aan lagu darin cabbirro.
Boggaan2 daqiiqo akhri
Dulmar
Waxay u oggolaadaan shabakadaha inay arkaan macnaha guud, oo muhiim u ah kala qaybinta iyo maqalka, iyadoo la ilaalinayo xallinta.
quusid qoto dheer
Kernel-ka kontorolka caadiga ah wuxuu taabtaa pixels ku xiga. Isku-dhafka fidsan wuxuu ku kala qaadaa isla miisaanka kernel-ka marka loo eego heerka fidinta, isaga oo ka boodaya pixels inta u dhaxaysa, markaa 3x3 kernel oo leh fidin 2 ayaa ku fidsan gobolka 5x5 iyadoo weli la isticmaalayo 9 miisaan. Tani waxay si xad dhaaf ah u ballaarineysaa goobta soo dhaweynta marka aad isku dhejiso lakabyo leh heerar sii kordhaya, taasoo u ogolaaneysa shabakadu inay isu geyso macnaha guud iyada oo aan la isku darin ama la jiidin taasoo lumin doonta khariidadda astaanta. Erayga atrous wuxuu ka yimid Faransiiska oo ah surwaal, oo macnaheedu yahay godad. Tani waa mid aad u qiimo badan marka loo eego hawlaha saadaasha cufan sida qaybinta semantic, halkaas oo aad u baahan tahay labadaba muuqaal ballaadhan iyo wax soo saar sax ah oo pixel ah, iyo WaveNet si aad u qaabeyso ku tiirsanaanta maqalka dheer.
Saamaynta Istiraatijiyadeed
Qiimaha iyo miisaaniyada
Go'aamada qaab-dhismeedku waxay horseedaan waxqabadka iyo kharashka hawlgalka sannadaha.
Go'aamo cad
Waxbarashada farsamada waxay ka caawisaa kooxaha inay doortaan xidhmo sax ah, ma aha oo kaliya kan ugu cusub.
Xakamaynta tayada
Doorashooyinka injineernimada ee wanaagsan waxay yareeyaan shilalka la isku halleyn karo ee wax soo saarka.
Mustaqbalka Kacdoonnada Dilaac iyo Argagaxa leh
Isku-dhafan la kala furfuray ayaa udub dhexaad u ah qaybta semantic iyo panoptic, sawirka caafimaadka, iyo jiilka maqalka. Waxay si sii kordheysa ugu dhex milmeen feejignaan, halkaas oo fidintu ay bixiso beero soo dhawayneed oo raqiis ah oo kaabaya is-fiirsiga. Cilmi-baadhistu waxay ku socotaa heerarka la-qabsiga iyo kuwa la baran karo iyo ka fogaanshiyaha waxyaabaha artifacts. Ka filo iyaga qaab hufan oo isku xigxiga dheer iyo fahamka goobta waqtiga-dhabta ah ee nidaamyada ismaamulka.
Dhaqangelinta Adduunka-dhabta ah
DeepLab waxay u isticmaashaa rogrogayo qallafsan iyo ASPP si ay u kala qaybiso semantic-ka ugu casrisan ee muuqaallada waddooyinka
WaveNet waxay soo ururisay is-waydaarsi sababayneed si ay u dhaliso maqal iyo hadal cayriin oo dhab ah
Kala qaybinta sawirka caafimaadka, sida xudduudaha burooyinka ama xubnaha, halkaas oo macnaha guud iyo tafatirka wanaagsan labaduba ay yihiin
Goobta-waqtiga-dhabta ah ee lagu falanqeynayo aragtida is-wadidda ee u baahan goobo waaweyn oo soo dhaweyn ah iyada oo aan lumin xallinta
Khatarta & Dariiqyada Ilaalada
Hagaajinta hal bartilmaameed waxay qarin kartaa daciifnimada nidaamka ballaaran.
Kaabayaasha dhaqaalaha iyo dayactirka inta badan waa la dhayalsadaa.
Nabadgelyada iyo daldaloolada u fiirsashada ayaa kori kara marka nidaamyadu noqdaan kuwo aad u adag.
Qorshe Hawleedka Dhaqangelinta
Qeex daahida, tayada, iyo bartilmaameedyada qiimaha ka hor inta aan la hirgelin.
Benchmark marka la eego culeyska dhabta ah iyo xaaladaha xogta.
La socodka qalabka khaladaadka, leexashada, iyo saamaynta isticmaalaha.
U diyaari dib-u-noqoshada iyo dariiqyada jawaab-celinta dhacdada ka hor inta aanad miisaan.
Sii wad Sahaminta
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 Dilated and Atrous Convolutions 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
Su'aalaha soo noqnoqda
Waa maxay Dilated and Arous Convolutions?
Isku-dhafka kala-goysyada (sidoo kale loo yaqaan convolutions atrous) waxay gelisaa farqiga u dhexeeya miisaanka shaandhada si kernelku u daboosho aag aad u weyn iyada oo aan lagu darin cabbirro. Waxay u oggolaadaan shabakadaha inay arkaan macnaha guud, oo muhiim u ah kala qaybinta iyo maqalka, iyadoo xallinta sidiisa loo ilaalinayo.
Muxuu heerka fidinta ka weyn 1 ku sameeyaa kernel convolution?
Kala-badhsiiya miisaannada jira, iyaga oo balaadhinaya goobta soo dhawaynta iyada oo aan la kordhin tirada cabbirrada.
3x3 kernel oo leh xadiga fidinta ee 2 ayaa si wax ku ool ah u cabbiraya meeqa gobol?
Dilation 2 waxaa jira hal boos oo la booday oo u dhexeeya miisaan kasta, markaa kernelku wuxuu gaaraa aag 5x5 iyadoo weli la isticmaalayo 9 miisaan.
Waa maxay sababta isfaham la ballaadhinayo ay si gaar ah faa'iido ugu leeyihiin qaybinta semantiga?
Qaybinta waxay u baahan tahay macnaha guud iyo tafaasiisha heerka pixel hal mar; fiditaanka ayaa koraysa macnaha guud iyada oo aan la yarayn in isku-darka ama socodka ay keeni karto.
Muxuu ka yimid ereyga atrous?
Atrous wuxuu ka yimid Faransiiska surwaal, oo leh godad, qeexaya daldaloolada la geliyo miisaanka kernel-ka.
WaveNet dhexdeeda, maxay isu-garabyada kala-baxa ah ugu qiimo badan yihiin?
Si badheedh ah oo kor u kaca heerarka fidinta waxay u ogolaataa qaabka WaveNet ku tiirsanaanta kumanaan muunado maqal ah oo aan lahayn kernels waaweyn.
Sii wad waxbarashada
Tilmaamaha la xidhiidha
Tilmaamayaal badan ayaa loo doortay mawduucan