Member-only story
Face generation with simple conditional Probability Model
It’s amazing to find simple mean of faces can generate a new face, it might not be the perfect face but shows the power of good labeled data. Below are the few very simple steps to try this out and experiment with generating new faces with different face attributes
Lets’ generate faces by sampling from a statistical model without even using a neural network.
Import Libraries
Download CelebFaces Attributes Dataset (CelebA) TensorFlow dataset
CelebFaces Attributes Dataset (CelebA) is a large-scale face attributes dataset with more than 200K celebrity images, each with 40 attribute annotations. The images in this dataset cover large pose variations and background clutter. CelebA has large diversities, large quantities, and rich annotations, including — 10,177 number of identities, — 202,599 number of face images, and — 5 landmark locations, 40 binary attributes annotations per image. Below is the link of CelebA dataset, if want to know more about dataset.
Preview some example images
Lets look at some sample images in dataset.
