CycleGAN: Tafsiri ya Picha-kwa-Picha isiyo na Jozi yenye Utaratibu wa Mzunguko
CycleGAN, iliyoanzishwa na Zhu et al. katika ICCV 2017, hujifunza kutafsiri picha kati ya nyanja mbili za kuona bila kuhitaji mifano ya mafunzo yenye jozi. Inafunza vizalishi viwili na wagunduzi wawili kwa wakati mmoja, ikilazimisha kizuizi cha utaratibu wa mzunguko ili picha iliyotafsiriwa kutoka nyanja X hadi Y na kurudi tena irudishe ile ya awali. Hii huifanya ifae wakati wowote ambapo seti kubwa za data zilizolingana hazipatikani, kama vile kubadilisha picha za kupiga picha kuwa mitindo ya sanaa, kubadilisha mandhari ya kiangazi kuwa mandhari ya majira ya baridi, au kuweka picha za setilaiti kwenye vigae vya ramani.
Soma mbinu kamili
Ingia kwa akaunti ya bure ili kusoma sehemu hii.
Method map
The neighbourhood of related methods — select a node to explore.
Vyanzo
- Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017). Unpaired image-to-image translation using cycle-consistent adversarial networks. IEEE International Conference on Computer Vision (ICCV), 2242–2251. DOI: 10.1109/ICCV.2017.244 ↗
Jinsi ya kunukuu ukurasa huu
ScholarGate. (2026, June 2). CycleGAN (Cycle-Consistent Image Translation). ScholarGate. https://scholargate.app/sw/deep-learning/cyclegan
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Mtandao wa Kushawishi unaozalisha (Generative Adversarial Network - GAN)Ujifunzaji wa Kina↔ compare
- Usanifu wa Mtindo wa NeuralUjifunzaji wa Kina↔ compare
- Wasserstein GAN (WGAN)Ujifunzaji wa Kina↔ compare
Imerejelewa na
Umeona tatizo kwenye ukurasa huu? Ripoti au pendekeza marekebisho →