Using generative adversarial networks to generate new images of faces (datasets: MNIST, CelebA).
-
Updated
Aug 1, 2017 - HTML
Using generative adversarial networks to generate new images of faces (datasets: MNIST, CelebA).
In this project, I’ve used Generative Adversarial Networks (GANs) to generate new images of human faces from scratch, based on the neural networks being trained on real human faces. I used the MNIST dataset and CelebFaces Attributes (CelebA) dataset in this project.
My solution to the Udacity project of creating a DCGAN to create new face images based on the CelebA image database
Use GANs with normalization techniques like dropouts, batch normalization along with having a low variance in kernel weight initialization, achieve realistic images of faces trained on the CelebA dataset. Images also have been generated of hand written digits after being trained on the MNIST dataset. This would be useful for generating training …
Add a description, image, and links to the celeba-dataset topic page so that developers can more easily learn about it.
To associate your repository with the celeba-dataset topic, visit your repo's landing page and select "manage topics."