Kayayyakin AI JAGORA

CogVideo da kuma CogVideoX

CogVideo (2022) shine farkon babban sikelin buɗaɗɗen rubutu-zuwa-bidiyo samfurin, kuma CogVideoX (2024) shine mafi kyawun magajinsa na buɗe tushen daga Tsinghua/Zhipu AI.

2 min karatuAn sabunta ta ƙarshe

Dubawa

They matter because they put high-quality video generation into the hands of the open community, not just big corporate labs.

Zurfafa nutsewa

CogVideo, wanda aka saki a cikin 2022, an gina shi akan mai canza rubutu-zuwa Hoto na CogView2 kuma yayi amfani da tsarin ƙima mai yawa, tsarin kai tsaye don samar da gajerun shirye-shiryen bidiyo, ya zama farkon fitowar babban samfurin rubutu-zuwa-bidiyo da goyan bayan Sinanci da Ingilishi. Magajin sa na 2024, CogVideoX, cikakke ne na sake fasalin: yana amfani da 3D causal bambance-bambancen autoencoder don matsa bidiyo a cikin sarari da lokaci, sannan ƙwararren Mai Canjawa tare da maƙasudin yadawa wanda ke halarta tare da rubutu da alamun bidiyo tare. Samfuran CogVideoX (a cikin masu girma kamar 2B da sigogi na 5B) suna haifar da daƙiƙa da yawa na daidaituwa, bidiyo mai motsi a ƙuduri kamar 720x480 da goyan bayan hoto-zuwa-bidiyo da ci gaba da bidiyo. Mahimmanci, ma'auni da lambobi jama'a ne, suna haɓaka ɗimbin waƙoƙi, kayan aiki, da bincike na al'umma.

Fahimtar Fasaha

CogVideoX's 3D causal VAE yana raguwa da ɗanyen bidiyo zuwa ƙaramin ƙarami mai ɓoyewa, yana yanke ƙididdige ƙididdigewa ta yadda mai canzawa zai iya yin ƙira mai tsayi mai araha. ƙwararren Mai Canjawa yana aiwatar da ka'idar daidaitawa kuma yana haɗa rubutu da alamu na gani don haka hanyoyin biyu su halarci juna kai tsaye, suna haɓaka daidaitawar rubutu-bidiyo. Horowar ci gaba akan ƙara ƙuduri da tsawon lokaci, tare da taka tsantsan bayanan bayanai, yana haifar da santsi, mafi aminci motsi na ma'ana.

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 CogVideo da CogVideoX

A matsayin ɗaya daga cikin samfuran bidiyo mafi ƙarfi da aka buɗe, CogVideoX yana ƙulla tsarin yanayin yanayi mai saurin girma na sauti mai kyau, adaftar sarrafawa, da tsawan lokaci mai tsawo. Yi tsammanin ci gaba da samun riba a cikin tsayin faifan bidiyo, ƙuduri, gaskiyar motsi, da iya sarrafawa, da ƙarin haɗa kai tare da hoto-zuwa-bidiyo da gyaran ayyukan aiki. Ma'aunin nauyi na buɗewa yana nufin ƙungiyoyin sa-kai, masu bincike, da ƙananan ɗakunan karatu na iya ginawa akan tsarar bidiyo na aji na kan iyaka ba tare da kiyaye ƙofa ta mallaka ba, haɓaka duka biyun ƙirƙira da gwajin mai da hankali kan aminci.

Aiwatar da Gaskiyar Duniya

Ƙirƙirar ɗan gajeren shirin labari daga faɗakarwar Sinanci ko Ingilishi ta amfani da cikakken ma'aunin nauyi

Juya hoton da aka ɗorawa guda ɗaya zuwa bidiyo mai motsi ta hanyar CogVideoX hoto-zuwa-bidiyo

Kyakkyawan daidaita samfurin buɗaɗɗe akan salo na al'ada ko hali don motsin indie

Masu bincike suna nuna sabbin hanyoyin samar da bidiyo akan tushen buɗaɗɗen da za'a iya maimaitawa

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

1

Ƙayyade ma'auni na karɓa don daidaito, tunowa, da farashi na kuskure.

2

Gwada tare da bayanan da suka dace da ainihin yanayin samarwa.

3

Ƙara bita na ɗan adam don ƙarancin amincewa ko tsinkaya mai tasiri.

4

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 CogVideo and CogVideoX quiz

Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.

Fara tambayoyi

Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation

Jagora na gaba

InstructPix2Pix Umarnin Gyara

Tambayoyin da ake yawan yi

What is CogVideo and CogVideoX?

CogVideo (2022) shine farkon babban sikelin buɗaɗɗen rubutu-zuwa-bidiyo samfurin, kuma CogVideoX (2024) shine mafi kyawun magajinsa na buɗe tushen daga Tsinghua/Zhipu AI. Suna da mahimmanci saboda suna sanya tsararrun bidiyo masu inganci a hannun jama'a masu buɗe ido, ba kawai manyan labs na kamfanoni ba.

Menene sananne game da fitowar CogVideo 2022?

CogVideo shine babban samfurin rubutu-zuwa-bidiyo na farko da aka fitar a bayyane, yana tallafawa duka Sinanci da Ingilishi.

Menene dalilin 3D na CogVideoX VAE ya cim ma?

3D causal VAE yana damfara bidiyo a duk faɗin sararin samaniya da na ɗan lokaci, yana rage ƙididdige ƙididdigewa don taswira zai iya ƙirƙira shi da kyau.

Ta yaya ƙwararren Mai Canjawa a CogVideoX ke sarrafa rubutu da bidiyo?

Kwararrun Mai Canjawa yana haɗa rubutu da alamu na gani tare da daidaitawa na daidaitawa, haɓaka jeri tsakanin gaggawa da ƙirƙirar bidiyo.

Wane rukuni ne suka haɓaka CogVideo da CogVideoX?

Iyalin CogVideo sun fito ne daga masu binciken da ke da alaƙa da Jami'ar Tsinghua da Zhipu AI.

Bayan rubutu-zuwa-bidiyo, wane ƙarin ƙarfi CogVideoX ke tallafawa?

CogVideoX na iya rayar da hoto mai tsayayye (hoto-zuwa-bidiyo) kuma ya tsawaita shirye-shiryen da ake da su (ci gaba), sama da faɗakarwar rubutu.