Gane Aiki
Fahimtar aiki shine aikin koyar da kwamfutoci don gano abin da mutane ko abubuwa * suke yi * a cikin bidiyo - gudu, daga hannu, faɗuwa, buɗe kofa - ba kawai abin da ke bayyana a cikin firam ɗaya ba.
Dubawa
It matters because understanding motion over time unlocks applications from sports analytics to elderly fall detection.
Zurfafa nutsewa
Gane aikin ya wuce tsayayyen hoto ta hanyar tunani game da yadda pixels ke canzawa cikin lokaci. Firam ɗaya na iya nuna mutum tsakiyar iska; jeri ne kawai ke bayyana ko suna tsalle-tsalle, suna fadowa, ko kuma suna nutsewa. Tsarin farko na fasalulluka motsin hannu kamar kwararar gani da ɗumbin hanyoyi. Hanyoyi na zamani suna amfani da hanyoyin sadarwa masu zurfi: tsarin tsarin gine-gine guda biyu (firam ɗin RGB) da motsi (gudanar gani) daban; 3D convolutional networks (kamar C3D da I3D) zamewa tacewa ta sararin samaniya * da * lokaci; da masu sauya bidiyo (TimeSformer, VideoMAE) suna amfani da hankali a cikin facin lokaci-lokaci. Madaidaitan ma'auni sun haɗa da Kinetics (darussan aikin ɗan adam 700 daga YouTube), UCF101, da Wani abu-Wani abu, wanda ke tilasta ƙira don fahimtar alkiblar ɗan lokaci maimakon mahallin yanayi kawai.
Fahimtar Fasaha
Babban ƙalubalen shine ƙirƙira girman ɗan lokaci. Juyin Juya Halin 3D yana faɗaɗa matattar 2D ta al'ada tare da zurfin axis mai faɗin firam da yawa, don haka yana koyon tsarin motsi kai tsaye. Dabarar ta I3D tana 'ɗaɗa' ma'aunin nauyi daga hanyar sadarwar hoto na 2D da aka riga aka horar akan ImageNet zuwa cikin 3D ta hanyar maimaita su a tsawon lokaci, yana ba da madaidaicin farawa. Hanyoyi guda biyu maimakon haka ciyar da kwararar gani da aka ƙididdige su zuwa wani reshe na daban, ɓoye motsi a sarari sannan a haɗa shi tare da fasalulluka.
Dabarun Tasiri
Gudu da sikelin
Kayayyakin AI na iya sarrafa aiki da bincike, ganowa, da ayyuka masu alama a sikelin.
Gina zaɓuɓɓuka
Ƙungiyoyin ƙirƙira za su iya samar da ra'ayoyi cikin sauri tare da ƙarancin bita da hannu.
Ƙungiya da aikin aiki
Ayyuka na iya amfani da siginar hoto da bidiyo waɗanda a baya suke da wahalar aiwatarwa.
Makomar Aiki Ganewa
Filin yana jujjuya zuwa ga ingantattun na'urorin taswirar bidiyo da horarwa na kulawa da kai (mased video modeling) waɗanda ke koyo daga fim ɗin da ba a lakafta su ba, yanke dogaro ga bayanai masu tsada. Yi tsammanin haɗin kai tare da ƙirar harsunan multimodal don haka tsarin ba zai iya yin lakabin ayyuka kawai ba amma bayyana da kuma dalili game da su cikin yare na halitta. Ganewar lokaci na gaske, kan na'urar don sawa, injiniyoyin mutum-mutumi, da kyamarori masu wayo shine babban kan iyaka, tare da ingantaccen ƙwarewa wanda ke bambance dabara, motsi iri ɗaya.
Aiwatar da Gaskiyar Duniya
Tsarin gano faɗuwa a cikin gidajen kula da tsofaffi waɗanda ke faɗakar da ma'aikata lokacin da mazaunin gida ya faɗi, yana bambanta faɗuwa daga zama ko kwance.
Dabarun nazarin wasanni waɗanda ke yiwa hidima ta atomatik, takalmi, da harbi a cikin faifan wasa don koyawa da abubuwan watsa shirye-shirye
Sa ido da sa ido kan tsaro wanda ke nuna halayen da ba a saba ba kamar fada, ja da baya, ko wani ya hau shinge
Abubuwan musaya masu sarrafa motsin motsi da ƙa'idodin motsa jiki waɗanda ke ƙidayar maimaitawa da duba sigar motsa jiki ta hanyar gane motsin jiki akan lokaci.
Hatsari & Tsare-tsare
Haƙƙoƙin hoto da yarda na iya zama haxarin doka idan ba a fayyace ba.
Ayyukan samfuri na iya bambanta a ko'ina cikin haske, ƙididdiga, da mahalli.
Ƙarya tabbataccen ƙila ba za a iya lura da shi ba sai dai idan an kula da ƙofofin amincewa.
Taswirar Hanya
Ƙayyade ma'auni na karɓa don daidaito, tunowa, da farashi na kuskure.
Gwada tare da bayanan da suka dace da ainihin yanayin samarwa.
Ƙara bita na ɗan adam don ƙarancin amincewa ko tsinkaya mai tasiri.
Bi diddigin ƙirar ƙira kuma sake ingantawa bayan canje-canjen kamara ko saitin bayanai.
Ci gaba da Bincike
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 Action Recognition 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
Jagora na gaba
Gane Fuska
Tambayoyin da ake yawan yi
What is Action Recognition?
Fahimtar aiki shine aikin koyar da kwamfutoci don gano abin da mutane ko abubuwa * suke yi * a cikin bidiyo - gudu, daga hannu, faɗuwa, buɗe kofa - ba kawai abin da ke bayyana a cikin firam ɗaya ba. Yana da mahimmanci saboda fahimtar motsi akan lokaci yana buɗe aikace-aikace daga nazarin wasanni zuwa gano faɗuwar tsofaffi.
Menene ainihin bambance bambance-bambancen aiki da rarrabuwar hoto?
Samfurin gano aikin yana motsi akan jeri na firam ɗin, tunda har yanzu hoto ɗaya sau da yawa ba zai iya bayyana ko wani yana tsalle, faɗuwa, ko nutsewa ba.
A cikin hanyar sadarwa mai gane aikin rafi biyu na gargajiya, menene rafi na biyu yakan aiwatar?
Gine-gine masu rafi biyu suna amfani da reshe ɗaya don bayyanar (firam ɗin RGB) da reshe na biyu don kwararar gani, wanda ke bayyana motsi tsakanin firam ɗin.
Menene juyi na 3D ke ƙara idan aka kwatanta da daidaitaccen juyi na 2D?
Juyin 3D yana faɗaɗa tacewa tare da zurfin/lokaci mai girma, yana barin shi ya koyi tsarin motsi kai tsaye a cikin firam ɗin jere.
Menene dabarar 'kumburi' da samfurin I3D ke amfani dashi?
I3D yana haɓaka ma'aunin nauyi da aka riga aka horar akan hotunan 2D (kamar ImageNet) zuwa cikin 3D ta hanyar kwafa su tare da axis na ɗan lokaci, yana ba da ingantaccen farawa.
Me yasa saitin bayanan Wani abu-Wani abu sananne ne don gane aiki?
An ƙera wani abu-wani abu don aikin ya dogara da tsari na ɗan lokaci da motsi, yana hana samfura yin magudi ta amfani da bango ko bayyanar abu kaɗai.