Nchọpụta ihe mepere emepe
Nchọpụta ihe mepere emepe na-eme ka ihe atụ chọta na igbe ihe ederede aka ike kọwara, gụnyere edemede ọ hụbeghị ka akpọnyere aha n'oge ọzụzụ.
Nchịkọta
It matters because traditional detectors are locked to a fixed list of classes, while open-vocabulary models can detect almost anything you can name.
Ime miri emi
A zụrụ ihe nchọta oge ochie n'ụdị usoro mechiri emechi, kwuo klaasị 80 dị na COCO, ha enweghị ike ịmata 'ihe' na mpụga ndepụta ahụ. Nchọpụta okwu mepere emepe na-ebelata oke site na ịhazi atụmatụ mpaghara a na-ahụ anya na oghere ntinye asụsụ ọhụụ, nke a na-amụta site na nnukwu ihe abụọ ederede (dị ka ọ dị na CLIP). Na ntinye, ị na-ebunye akara ederede, ihe nlereanya ahụ na-etinye akara ndị ahụ, ọ dakọtara na mpaghara achọpụtara na nke ọ bụla ntinye ederede kacha nso, yabụ edemede akwụkwọ na-arụ ọrụ ma ọ bụrụhaala na ị nwere ike ịkọwa ha. Sistemu dị ka ViLD, GLIP, OWL-ViT, Detic, na Grounding DINO kwalitere ụzọ a site na ijikọta ọkpụkpụ nchọta na ntọ ala asụsụ yana site n'ịzụ nnukwu, aha na-adịghị ike ma ọ bụ na-akụda datasets.
Nghọta nka nka
Aghụghọ a bụ iji ntinye ederede dochie oyi akwa classifier edoziri. Kama ịmụta otu vector dị arọ kwa klaasị a ma ama, onye na-achọpụta ihe na-arụ ọrụ mpaghara ọ bụla n'otu oghere dị ka ihe ngbanwe asụsụ; Nkewa na-aghọ ntụnyere myirịta n'etiti njirimara mpaghara yana ntinye aha ma ọ bụ nkebiokwu nke onye ọrụ nyere. N'ihi na ederede ederede na-ejikọta ọnụ na okwu ndị a na-adịghị ahụ anya, ịgbanwee n'ime eriri akara ọhụrụ n'oge ule na-enyere aka ịchọpụta edemede na-anọghị na data ọzụzụ igbe bounding.
Mmetụta atụmatụ
Ọsọ na ọnụ ọgụgụ
Visual AI nwere ike megharịa nyocha, nchọpụta na mkpado ọrụ n'ọtụtụ.
Mee nhọrọ
Otu ndị na-emepụta ihe nwere ike imepụta echiche ngwa ngwa site na ngbanwe akwụkwọ ntuziaka ole na ole.
Team na usoro ọrụ
Ọrụ nwere ike iji onyonyo na akara vidiyo siri ike ịhazi.
Ọdịnihu nke nchọpụta ihe mepere emepe
Nchọpụta okwu mepere emepe na-ejikọta ya na ntọ ala na nkewa, ebe nkebi ahịrịokwu na-enweghị onwe (ọ bụghị naanị otu okwu) na-akọwapụta ihe, yana usoro ngwa ngwa jikọtara ya na ụdị dịka SAM maka masks. Na-atụ anya izi ezi-shot efu siri ike karị, ajụjụ ederede dị ogologo yana karịa ('ọka na-acha ọbara ọbara n'azụ laptọọpụ'), yana njikọ siri ike na ndị enyemaka multimodal na-achọpụta na achọrọ. Ka ọzụzụ onyonyo-ederede webụsaịtị na-akawanye mma, ahịrị dị n'etiti nchọpụta, iweghachite, na nghọta asụsụ ga-aga n'ihu na-edobe anya na ntọ ala n'ozuzu.
Mmejuputa n'ezie n'ụwa
Na-achọ onyonyo maka ihe ndị na-adịghị ahụkebe ma ọ bụ omenala site na ịpị aha ha na-enweghị ọzụzụ
Sistemụ roboti na-achọpụta ihe onye ọrụ na-akpọ aha n'asụsụ eke tupu ya aghọta ya
Ịkpọ aha ọdụ data na-akpaghị aka site na ịchọpụta ọtụtụ ụdị ọhụrụ site na ndetu ederede
Nhazi ọdịnaya nke ọkọlọtọ kọwapụtara ihe adịghị na akara izizi izizi
Ihe ize ndụ & okporo ụzọ nche
Ikike onyonyo na nkwenye nwere ike bụrụ ihe egwu dị n'iwu ma ọ bụrụ na edoghị anya.
Ọrụ nlereanya nwere ike ịdịgasị iche n'ofe ọkụ, igwe mmadụ, na gburugburu.
Enwere ike ghara ịhụ ihe dị mma ma ọ bụrụ na enyochaghị oke ntụkwasị obi.
Map mmejuputa
Kọwaa ụkpụrụ nnabata maka nkenke, icheta, na ụgwọ njehie.
Nwalee na data dabara na ọnọdụ mmepụta n'ezie.
Tinye nyocha mmadụ maka obere obi ike ma ọ bụ amụma mmetụta dị elu.
Sochie ihe nlere anya wee megharịa ka emechara mgbanwe igwefoto ma ọ bụ dataset.
Nọgide na-eme nchọpụta
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 Open-Vocabulary Object Detection 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
Ntuziaka na-esote
Nchọpụta ihe
Ajụjụ a na-ajụkarị
What is Open-Vocabulary Object Detection?
Nchọpụta ihe mepere emepe na-eme ka ihe atụ chọta na igbe ihe ederede aka ike kọwara, gụnyere edemede ọ hụbeghị ka akpọnyere aha n'oge ọzụzụ. Ọ dị mkpa n'ihi na akpọchiri ihe nchọpụta ọdịnala na ndepụta klaasị edobere, ebe ụdị okwu mepere emepe nwere ike ịchọpụta ihe ọ bụla ị nwere ike ịkpọ aha.
Gịnị bụ ike na-akọwapụta ihe oghe-okwu?
Nchọpụta okwu mepere emepe nwere ike wepụta ngalaba ederede akọwapụtara n'oge ule, ọbụlagodi ndị ọ na-ahụtụbeghị ka akara ya na igbe ejichi.
Kedu ka ụdị ndị a si ekewa mpaghara achọpụtara?
Ha na-emepụta mpaghara ka ọ bụrụ oghere ntinye asụsụ ọhụụ wee dakọtara mpaghara ọ bụla na ntinye akara ederede kacha nso.
Kedu akụrụngwa nkuzi izizi na-enye aka ịchọpụta okwu mepere emepe?
Nnukwu data ederede onyonyo (dị ka ụdị ụdị CLIP ji eme ya) na-ewulite oghere nkekọrịta nke na-eme ka ederede chịkọta ya na ụdị ọhụrụ.
Kedu akụrụngwa edochiri ma e jiri ya tụnyere ihe nchọta mechiri emechi?
Kama vectors arọ klaasị edobere, nhazi ọkwa na-eji myirịta na ntinye ederede, yabụ enwere ike ịgbakwunye eriri akara ọhụrụ n'oge ule.
Kedu n'ime ihe ndị a bụ ihe atụ mepere emepe ma ọ bụ ihe nchọta ala?
Grounding DINO, yana GLIP, OWL-ViT, ViLD, na Detic, bụ ndị ama ama nke ọma mepere emepe ma ọ bụ ihe nchọpụta ala.