Baby Face Generation with Generative Adversarial Neural Networks: A Case Study
[ X ]
Tarih
2020
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Generative Adversarial Networks (GANs) are increasingly applied to train generative models with neuralnetworks, especially in computer vision studies. Since being introduced in 2014, many image generationstudies incorporating GANs have demonstrated promising results for producing highly convincing fakeimages of animals, landscapes, and human faces. We build a GAN structure to generate realistic baby faceimages from a small data set of 673 color 200×200 pixel images obtained from a Kaggle data set by followingprevious studies that demonstrated how GANs could be used for image generation from a limited number oftraining samples. The reason we limit especially as baby faces is that we aim to achieve success with a limitednumber of training data. For evaluation, experiments and case studies are one of the most consideredtechniques. The results of this study help identify issues requiring further investigation in comment analysisresearch. In this context, we presented the loss values of the generator and discriminator during the trainingprocess. The discriminator losses are around of 0.7 and the generator is between 0.7 and 0.9. The high qualityimages are produced about 300th epochs.
Açıklama
Anahtar Kelimeler
Bilgisayar Bilimleri, Yapay Zeka
Kaynak
Acta Infologica
WoS Q Değeri
Scopus Q Değeri
Cilt
4
Sayı
1
