Ƙimar Zurfin Juna
Ƙididdiga mai zurfi na monocular yana annabta nisa kowane pixel daga hoto na yau da kullun - babu kyamarar sitiriyo, lidar, ko zurfin firikwensin da ake buƙata.
Dubawa
It lets one camera perceive 3D structure from a flat 2D image.
Zurfafa nutsewa
Mutane na iya yin hukunci mai zurfi daga ido ɗaya ta amfani da alamu kamar hangen nesa, girman dangi, gradients na rubutu, shading, da ɓoyewa. Ƙididdiga mai zurfi na monocular yana koyar da hanyoyin sadarwar jijiyoyi iri ɗaya: ciyarwa a cikin hoton RGB guda ɗaya da fitar da zurfin ƙimar kowane pixel. Saboda hoton 2D yana da ban sha'awa game da cikakken ma'auni, aikin yana da wahala - yawancin al'amuran 3D na iya aiwatar da hoto ɗaya. Cibiyoyin sadarwa suna koyon ilimin kididdiga kafin daga manyan bayanan bayanai don warware wannan. Horowa ya zo cikin dandano biyu: kulawa, ta amfani da zurfin-gaskiya daga lidar ko na'urori masu auna firikwensin RGB-D, da kuma kulawa da kai, wanda ke koyon zurfi zalla daga bidiyo ko nau'ikan sitiriyo ta hanyar tilasta cewa zurfin da aka annabta daidai ya mayar da ra'ayi ɗaya zuwa wani. Samfurin tushe na baya-bayan nan kamar MiDaS da Zurfin Duk wani abu yana ba da juzu'i a cikin al'amuran da ba a gani.
Fahimtar Fasaha
Hanyoyin kulawa da kai suna yin amfani da ilimin lissafi maimakon lakabi. Idan aka ba da ra'ayoyi biyu (firam ɗin bidiyo na sitiriyo ko jere) da taswirar zurfin tsinkaya tare da motsin kyamara, ƙirar tana ɗaukar hoto ɗaya don sake gina ɗayan; kuskuren sake gina matakin pixel ya zama siginar horo. Wannan hasara na 'view-synthesis' yana nufin za a iya koyan zurfin daga ɗanyen bidiyon da ba shi da lakabi. Ƙayyadaddun maɓalli shine shubuhar ma'auni: zurfin monocular sau da yawa daidai ne kawai har zuwa wanda ba a san shi ba sai dai in an daidaita shi da sanannen tunani ko kulawar awo.
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 Ƙimar Zurfin Monocular
Samfuran tushen tushen zurfin gaba ɗaya waɗanda aka horar akan miliyoyin hotuna masu gauraya suna matsawa zuwa ga abin dogaro, zurfin awo (ma'auni) a kowane fage, har ma waɗanda ba a taɓa gani ba a horo. Yi tsammanin haɗuwa mai ƙarfi tare da kwararar gani da SLAM don cikakken sake fasalin yanayin 3D, ƙirar haske waɗanda ke gudana kai tsaye akan wayoyi da naúrar kai, da ƙarfi mai ƙarfi. Wannan zai sa hasashe mai arziƙi mai arha kuma ya zama a ko'ina, ana samunsa daga kowace kyamara ɗaya maimakon ma'ajin zurfin fahimtar zurfin tunani.
Aiwatar da Gaskiyar Duniya
Yanayin hoto na wayo mai simintin blur baya (bokeh) ta kiyasin tazarar magana-da-baya
Haƙiƙa ƙa'idodin ƙa'idodi waɗanda ke sanya abubuwan kama-da-wane don haka su zauna daidai a bayan kayan daki na ainihi
Jiragen sama marasa matuki da mutummutumi masu rahusa suna guje wa cikas ta amfani da kyamarar gaba guda ɗaya
Canza hotuna da fina-finai 2D zuwa 3D ta hanyar fahimtar zurfin kowane pixel don nunin sitiriyo.
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 Monocular Depth Estimation 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
Ƙimar Zurfin Sitiriyo
Tambayoyin da ake yawan yi
What is Monocular Depth Estimation?
Ƙididdiga mai zurfi na monocular yana annabta nisa kowane pixel daga hoto na yau da kullun - babu kyamarar sitiriyo, lidar, ko zurfin firikwensin da ake buƙata. Yana ba da damar kyamara ɗaya ta gane tsarin 3D daga hoto mai faɗi 2D.
Menene ya sa kimanta zurfin 'monocular'?
'Monocular' na nufin ido/kamara ɗaya; Hanyar tana annabta zurfin daga hoton RGB guda ɗaya ba tare da sitiriyo ko na'urori masu zurfin aiki ba.
Me yasa kiyasin zurfin guda ɗaya yake da shubuha?
Hasashen 2D guda ɗaya ya rasa bayanin sikelin, don haka shirye-shiryen 3D da yawa sun yi daidai da hoto ɗaya; cibiyoyin sadarwa sun dogara da abubuwan da aka koya kafin su warware.
Ta yaya hanyoyin kulawa da kai za su koyi zurfi ba tare da alamun gaskiya ba?
Yin amfani da zurfin da aka annabta da motsin kyamara, ƙirar tana mayar da hoto ɗaya zuwa wani; Kuskuren photometric yana ba da siginar ilmantarwa, babu alamun da ake buƙata.
Wanne alama ce cibiyoyin sadarwa guda ɗaya BASA dogaro kai tsaye don zurfin?
Bambancin sitiriyo yana buƙatar kyamarori biyu; Hanyoyi guda ɗaya sun dogara da alamun hoto ɗaya kamar girman, hangen nesa, rubutu, da ɓoyewa.
Menene 'shakawar sikelin' a cikin zurfin monocular?
Ba tare da kulawar awo ko sanannen tunani ba, zurfin monocular yana ba da daidaitaccen tsarin dangi amma maras tabbas ma'auni.